BZPEERPapers, people, evidence and research paths
Menu

Discover

SearchFind papers, people, topics, institutions and opportunities from one place.GraphSee how papers, authors, topics and evidence connect.

Research tools

HomeContinue from the papers, people and topics that matter now.CollectionsTurn saved papers into reading lists, evidence maps and shareable context.ResearchReconstruct the path that led you from search to paper to question.CollaborationMove from a paper, topic or expert request into a focused thread.PublishCreate full-text research articles with metadata, formulas and versions.

Network

ExpertsFind specialists whose work and reviews explain why they can help.CommunitiesJoin research groups organized around papers, topics and shared evidence.

Opportunities

ListingsBrowse roles, calls and collaborations connected to real research context.

Account

Log inReturn to your saved papers, graph views and active threads.Create accountStart saving papers, building a research trail and joining teams.
Log inCreate account

Source author record

Thomas Brox

Thomas Brox appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record
Computer VisionMachine LearningNeural and Evolutionary ComputingRoboticsArtificial Intelligencemath.OCNeurons and Cognition

Catalog footprint

What is connected

43works
7topics
4close collaborators

Actions

Connect this record

Log in to claim
Open graphBrowse works

Research graph

See the researcher in context

Open full explorer

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

43 published item(s)

preprint2026arXiv

Eliciting associations between clinical variables from LLMs via comparison questions across populations

The training data of large language models (LLMs) comprises a wide range of biomedical literature, reflecting data from many different patient populations. We investigate how it might be possible to recover information on correlation and causal links between patient characteristics, as a key building block for medical decision making. To avoid the pitfalls of direct elicitation, we propose an approach based on structured comparison questions, specifically patient comparison triplet questions. This is combined with a statistical model for the LLM representation that provides estimates of correlations without access to activations or model internals. Intuitively, we consider how similarity decisions of LLMs based on a first variable are affected by providing information on a second variable for one of the patients being assessed. We then induce prompt-level environment shifts to obtain correlation estimates for different subpopulations, which enables an invariant causal prediction (ICP) approach to obtain conservative candidate parent links. We demonstrate the method in two clinical domains, chronic obstructive pulmonary disease (COPD) and multiple sclerosis (MS). Across prompted environments, the elicited correlations are smooth, stable, and clinically interpretable, yet vary in a statistically significant way that supports downstream invariance testing, such that ICP provides a small set of candidate invariant parent links. These results show that indirect elicitation via triplet comparisons can recover meaningful association structure from LLMs and offer a cautious route from implicit correlations to causal statements that are congruent with LLM answering patterns.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

The Surprising Effectiveness of Canonical Knowledge Distillation for Semantic Segmentation

Recent knowledge distillation (KD) methods for semantic segmentation introduce increasingly complex hand-crafted objectives, yet are typically evaluated under fixed iteration schedules. These objectives substantially increase per-iteration cost, meaning equal iteration counts do not correspond to equal training budgets. It is therefore unclear whether reported gains reflect stronger distillation signals or simply greater compute. We show that iteration-based comparisons are misleading: when wall-clock compute is matched, canonical logit- and feature-based KD outperform recent segmentation-specific methods. Under extended training, feature-based distillation achieves state-of-the-art ResNet-18 performance on Cityscapes and ADE20K. A PSPNet ResNet-18 student closely approaches its ResNet-101 teacher despite using only one quarter of the parameters, reaching 99% of the teacher's mIoU on Cityscapes (79.0 vs 79.8) and 92% on ADE20K. Our results challenge the prevailing assumption that KD for segmentation requires task-specific mechanisms and suggest that scaling, rather than complex hand-crafted objectives, should guide future method design.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

A Benchmark and a Baseline for Robust Multi-view Depth Estimation

Recent deep learning approaches for multi-view depth estimation are employed either in a depth-from-video or a multi-view stereo setting. Despite different settings, these approaches are technically similar: they correlate multiple source views with a keyview to estimate a depth map for the keyview. In this work, we introduce the Robust Multi-View Depth Benchmark that is built upon a set of public datasets and allows evaluation in both settings on data from different domains. We evaluate recent approaches and find imbalanced performances across domains. Further, we consider a third setting, where camera poses are available and the objective is to estimate the corresponding depth maps with their correct scale. We show that recent approaches do not generalize across datasets in this setting. This is because their cost volume output runs out of distribution. To resolve this, we present the Robust MVD Baseline model for multi-view depth estimation, which is built upon existing components but employs a novel scale augmentation procedure. It can be applied for robust multi-view depth estimation, independent of the target data. We provide code for the proposed benchmark and baseline model at https://github.com/lmb-freiburg/robustmvd.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Conditional Visual Servoing for Multi-Step Tasks

Visual Servoing has been effectively used to move a robot into specific target locations or to track a recorded demonstration. It does not require manual programming, but it is typically limited to settings where one demonstration maps to one environment state. We propose a modular approach to extend visual servoing to scenarios with multiple demonstration sequences. We call this conditional servoing, as we choose the next demonstration conditioned on the observation of the robot. This method presents an appealing strategy to tackle multi-step problems, as individual demonstrations can be combined flexibly into a control policy. We propose different selection functions and compare them on a shape-sorting task in simulation. With the reprojection error yielding the best overall results, we implement this selection function on a real robot and show the efficacy of the proposed conditional servoing. For videos of our experiments, please check out our project page: https://lmb.informatik.uni-freiburg.de/projects/conditional_servoing/

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Localized Vision-Language Matching for Open-vocabulary Object Detection

In this work, we propose an open-vocabulary object detection method that, based on image-caption pairs, learns to detect novel object classes along with a given set of known classes. It is a two-stage training approach that first uses a location-guided image-caption matching technique to learn class labels for both novel and known classes in a weakly-supervised manner and second specializes the model for the object detection task using known class annotations. We show that a simple language model fits better than a large contextualized language model for detecting novel objects. Moreover, we introduce a consistency-regularization technique to better exploit image-caption pair information. Our method compares favorably to existing open-vocabulary detection approaches while being data-efficient. Source code is available at https://github.com/lmb-freiburg/locov .

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Neural Architecture Search for Dense Prediction Tasks in Computer Vision

The success of deep learning in recent years has lead to a rising demand for neural network architecture engineering. As a consequence, neural architecture search (NAS), which aims at automatically designing neural network architectures in a data-driven manner rather than manually, has evolved as a popular field of research. With the advent of weight sharing strategies across architectures, NAS has become applicable to a much wider range of problems. In particular, there are now many publications for dense prediction tasks in computer vision that require pixel-level predictions, such as semantic segmentation or object detection. These tasks come with novel challenges, such as higher memory footprints due to high-resolution data, learning multi-scale representations, longer training times, and more complex and larger neural architectures. In this manuscript, we provide an overview of NAS for dense prediction tasks by elaborating on these novel challenges and surveying ways to address them to ease future research and application of existing methods to novel problems.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

NeuRL: Closed-form Inverse Reinforcement Learning for Neural Decoding

Current neural decoding methods typically aim at explaining behavior based on neural activity via supervised learning. However, since generally there is a strong connection between learning of subjects and their expectations on long-term rewards, we propose NeuRL, an inverse reinforcement learning approach that (1) extracts an intrinsic reward function from collected trajectories of a subject in closed form, (2) maps neural signals to this intrinsic reward to account for long-term dependencies in the behavior and (3) predicts the simulated behavior for unseen neural signals by extracting Q-values and the corresponding Boltzmann policy based on the intrinsic reward values for these unseen neural signals. We show that NeuRL leads to better generalization and improved decoding performance compared to supervised approaches. We study the behavior of rats in a response-preparation task and evaluate the performance of NeuRL within simulated inhibition and per-trial behavior prediction. By assigning clear functional roles to defined neuronal populations our approach offers a new interpretation tool for complex neuronal data with testable predictions. In per-trial behavior prediction, our approach furthermore improves accuracy by up to 15% compared to traditional methods.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Pixel-level Correspondence for Self-Supervised Learning from Video

While self-supervised learning has enabled effective representation learning in the absence of labels, for vision, video remains a relatively untapped source of supervision. To address this, we propose Pixel-level Correspondence (PiCo), a method for dense contrastive learning from video. By tracking points with optical flow, we obtain a correspondence map which can be used to match local features at different points in time. We validate PiCo on standard benchmarks, outperforming self-supervised baselines on multiple dense prediction tasks, without compromising performance on image classification.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Probing Contextual Diversity for Dense Out-of-Distribution Detection

Detection of out-of-distribution (OoD) samples in the context of image classification has recently become an area of interest and active study, along with the topic of uncertainty estimation, to which it is closely related. In this paper we explore the task of OoD segmentation, which has been studied less than its classification counterpart and presents additional challenges. Segmentation is a dense prediction task for which the model's outcome for each pixel depends on its surroundings. The receptive field and the reliance on context play a role for distinguishing different classes and, correspondingly, for spotting OoD entities. We introduce MOoSe, an efficient strategy to leverage the various levels of context represented within semantic segmentation models and show that even a simple aggregation of multi-scale representations has consistently positive effects on OoD detection and uncertainty estimation.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Ranking Info Noise Contrastive Estimation: Boosting Contrastive Learning via Ranked Positives

This paper introduces Ranking Info Noise Contrastive Estimation (RINCE), a new member in the family of InfoNCE losses that preserves a ranked ordering of positive samples. In contrast to the standard InfoNCE loss, which requires a strict binary separation of the training pairs into similar and dissimilar samples, RINCE can exploit information about a similarity ranking for learning a corresponding embedding space. We show that the proposed loss function learns favorable embeddings compared to the standard InfoNCE whenever at least noisy ranking information can be obtained or when the definition of positives and negatives is blurry. We demonstrate this for a supervised classification task with additional superclass labels and noisy similarity scores. Furthermore, we show that RINCE can also be applied to unsupervised training with experiments on unsupervised representation learning from videos. In particular, the embedding yields higher classification accuracy, retrieval rates and performs better in out-of-distribution detection than the standard InfoNCE loss.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

RobotIO: A Python Library for Robot Manipulation Experiments

Setting up robot environments to quickly test newly developed algorithms is still a difficult and time consuming process. This presents a significant hurdle to researchers interested in performing real-world robotic experiments. RobotIO is a python library designed to solve this problem. It focuses on providing common, simple, and well structured python interfaces for robots, grippers, and cameras, etc. These are provided with implementations of these interfaces for common hardware. This enables code using RobotIO to be portable across different robot setups. In terms of architecture, RobotIO is designed to be compatible with OpenAI gym environments, as well as ROS; examples of both of these are provided. The library comes together with a number of helpful tools, such as camera calibration scripts and episode recording functionality that further support algorithm development.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Towards Total Recall in Industrial Anomaly Detection

Being able to spot defective parts is a critical component in large-scale industrial manufacturing. A particular challenge that we address in this work is the cold-start problem: fit a model using nominal (non-defective) example images only. While handcrafted solutions per class are possible, the goal is to build systems that work well simultaneously on many different tasks automatically. The best performing approaches combine embeddings from ImageNet models with an outlier detection model. In this paper, we extend on this line of work and propose \textbf{PatchCore}, which uses a maximally representative memory bank of nominal patch-features. PatchCore offers competitive inference times while achieving state-of-the-art performance for both detection and localization. On the challenging, widely used MVTec AD benchmark PatchCore achieves an image-level anomaly detection AUROC score of up to $99.6\%$, more than halving the error compared to the next best competitor. We further report competitive results on two additional datasets and also find competitive results in the few samples regime.\freefootnote{$^*$ Work done during a research internship at Amazon AWS.} Code: github.com/amazon-research/patchcore-inspection.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Towards Understanding Adversarial Robustness of Optical Flow Networks

Recent work demonstrated the lack of robustness of optical flow networks to physical patch-based adversarial attacks. The possibility to physically attack a basic component of automotive systems is a reason for serious concerns. In this paper, we analyze the cause of the problem and show that the lack of robustness is rooted in the classical aperture problem of optical flow estimation in combination with bad choices in the details of the network architecture. We show how these mistakes can be rectified in order to make optical flow networks robust to physical patch-based attacks. Additionally, we take a look at global white-box attacks in the scope of optical flow. We find that targeted white-box attacks can be crafted to bias flow estimation models towards any desired output, but this requires access to the input images and model weights. However, in the case of universal attacks, we find that optical flow networks are robust. Code is available at https://github.com/lmb-freiburg/understanding_flow_robustness.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Essentials for Class Incremental Learning

Contemporary neural networks are limited in their ability to learn from evolving streams of training data. When trained sequentially on new or evolving tasks, their accuracy drops sharply, making them unsuitable for many real-world applications. In this work, we shed light on the causes of this well-known yet unsolved phenomenon - often referred to as catastrophic forgetting - in a class-incremental setup. We show that a combination of simple components and a loss that balances intra-task and inter-task learning can already resolve forgetting to the same extent as more complex measures proposed in literature. Moreover, we identify poor quality of the learned representation as another reason for catastrophic forgetting in class-IL. We show that performance is correlated with secondary class information (dark knowledge) learned by the model and it can be improved by an appropriate regularizer. With these lessons learned, class-incremental learning results on CIFAR-100 and ImageNet improve over the state-of-the-art by a large margin, while keeping the approach simple.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

On Exposing the Challenging Long Tail in Future Prediction of Traffic Actors

Predicting the states of dynamic traffic actors into the future is important for autonomous systems to operate safelyand efficiently. Remarkably, the most critical scenarios aremuch less frequent and more complex than the uncriticalones. Therefore, uncritical cases dominate the prediction. In this paper, we address specifically the challenging scenarios at the long tail of the dataset distribution. Our analysis shows that the common losses tend to place challenging cases suboptimally in the embedding space. As a consequence, we propose to supplement the usual loss with aloss that places challenging cases closer to each other. This triggers sharing information among challenging cases andlearning specific predictive features. We show on four public datasets that this leads to improved performance on the challenging scenarios while the overall performance stays stable. The approach is agnostic w.r.t. the used network architecture, input modality or viewpoint, and can be integrated into existing solutions easily. Code is available at https://github.com/lmb-freiburg/Contrastive-Future-Trajectory-Prediction

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Adaptive Curriculum Generation from Demonstrations for Sim-to-Real Visuomotor Control

We propose Adaptive Curriculum Generation from Demonstrations (ACGD) for reinforcement learning in the presence of sparse rewards. Rather than designing shaped reward functions, ACGD adaptively sets the appropriate task difficulty for the learner by controlling where to sample from the demonstration trajectories and which set of simulation parameters to use. We show that training vision-based control policies in simulation while gradually increasing the difficulty of the task via ACGD improves the policy transfer to the real world. The degree of domain randomization is also gradually increased through the task difficulty. We demonstrate zero-shot transfer for two real-world manipulation tasks: pick-and-stow and block stacking. A video showing the results can be found at https://lmb.informatik.uni-freiburg.de/projects/curriculum/

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Beyond Single Stage Encoder-Decoder Networks: Deep Decoders for Semantic Image Segmentation

Single encoder-decoder methodologies for semantic segmentation are reaching their peak in terms of segmentation quality and efficiency per number of layers. To address these limitations, we propose a new architecture based on a decoder which uses a set of shallow networks for capturing more information content. The new decoder has a new topology of skip connections, namely backward and stacked residual connections. In order to further improve the architecture we introduce a weight function which aims to re-balance classes to increase the attention of the networks to under-represented objects. We carried out an extensive set of experiments that yielded state-of-the-art results for the CamVid, Gatech and Freiburg Forest datasets. Moreover, to further prove the effectiveness of our decoder, we conducted a set of experiments studying the impact of our decoder to state-of-the-art segmentation techniques. Additionally, we present a set of experiments augmenting semantic segmentation with optical flow information, showing that motion clues can boost pure image based semantic segmentation approaches.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

FlowControl: Optical Flow Based Visual Servoing

One-shot imitation is the vision of robot programming from a single demonstration, rather than by tedious construction of computer code. We present a practical method for realizing one-shot imitation for manipulation tasks, exploiting modern learning-based optical flow to perform real-time visual servoing. Our approach, which we call FlowControl, continuously tracks a demonstration video, using a specified foreground mask to attend to an object of interest. Using RGB-D observations, FlowControl requires no 3D object models, and is easy to set up. FlowControl inherits great robustness to visual appearance from decades of work in optical flow. We exhibit FlowControl on a range of problems, including ones requiring very precise motions, and ones requiring the ability to generalize.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Multimodal Future Localization and Emergence Prediction for Objects in Egocentric View with a Reachability Prior

In this paper, we investigate the problem of anticipating future dynamics, particularly the future location of other vehicles and pedestrians, in the view of a moving vehicle. We approach two fundamental challenges: (1) the partial visibility due to the egocentric view with a single RGB camera and considerable field-of-view change due to the egomotion of the vehicle; (2) the multimodality of the distribution of future states. In contrast to many previous works, we do not assume structural knowledge from maps. We rather estimate a reachability prior for certain classes of objects from the semantic map of the present image and propagate it into the future using the planned egomotion. Experiments show that the reachability prior combined with multi-hypotheses learning improves multimodal prediction of the future location of tracked objects and, for the first time, the emergence of new objects. We also demonstrate promising zero-shot transfer to unseen datasets. Source code is available at $\href{https://github.com/lmb-freiburg/FLN-EPN-RPN}{\text{this https URL.}}$

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Optimized Generic Feature Learning for Few-shot Classification across Domains

To learn models or features that generalize across tasks and domains is one of the grand goals of machine learning. In this paper, we propose to use cross-domain, cross-task data as validation objective for hyper-parameter optimization (HPO) to improve on this goal. Given a rich enough search space, optimization of hyper-parameters learn features that maximize validation performance and, due to the objective, generalize across tasks and domains. We demonstrate the effectiveness of this strategy on few-shot image classification within and across domains. The learned features outperform all previous few-shot and meta-learning approaches.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Overcoming Limitations of Mixture Density Networks: A Sampling and Fitting Framework for Multimodal Future Prediction

Future prediction is a fundamental principle of intelligence that helps plan actions and avoid possible dangers. As the future is uncertain to a large extent, modeling the uncertainty and multimodality of the future states is of great relevance. Existing approaches are rather limited in this regard and mostly yield a single hypothesis of the future or, at the best, strongly constrained mixture components that suffer from instabilities in training and mode collapse. In this work, we present an approach that involves the prediction of several samples of the future with a winner-takes-all loss and iterative grouping of samples to multiple modes. Moreover, we discuss how to evaluate predicted multimodal distributions, including the common real scenario, where only a single sample from the ground-truth distribution is available for evaluation. We show on synthetic and real data that the proposed approach triggers good estimates of multimodal distributions and avoids mode collapse. Source code is available at $\href{https://github.com/lmb-freiburg/Multimodal-Future-Prediction}{\text{this https URL.}}$

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Scaling Imitation Learning in Minecraft

Imitation learning is a powerful family of techniques for learning sensorimotor coordination in immersive environments. We apply imitation learning to attain state-of-the-art performance on hard exploration problems in the Minecraft environment. We report experiments that highlight the influence of network architecture, loss function, and data augmentation. An early version of our approach reached second place in the MineRL competition at NeurIPS 2019. Here we report stronger results that can be used as a starting point for future competition entries and related research. Our code is available at https://github.com/amiranas/minerl_imitation_learning.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Understanding and Robustifying Differentiable Architecture Search

Differentiable Architecture Search (DARTS) has attracted a lot of attention due to its simplicity and small search costs achieved by a continuous relaxation and an approximation of the resulting bi-level optimization problem. However, DARTS does not work robustly for new problems: we identify a wide range of search spaces for which DARTS yields degenerate architectures with very poor test performance. We study this failure mode and show that, while DARTS successfully minimizes validation loss, the found solutions generalize poorly when they coincide with high validation loss curvature in the architecture space. We show that by adding one of various types of regularization we can robustify DARTS to find solutions with less curvature and better generalization properties. Based on these observations, we propose several simple variations of DARTS that perform substantially more robustly in practice. Our observations are robust across five search spaces on three image classification tasks and also hold for the very different domains of disparity estimation (a dense regression task) and language modelling.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation

This paper introduces a network for volumetric segmentation that learns from sparsely annotated volumetric images. We outline two attractive use cases of this method: (1) In a semi-automated setup, the user annotates some slices in the volume to be segmented. The network learns from these sparse annotations and provides a dense 3D segmentation. (2) In a fully-automated setup, we assume that a representative, sparsely annotated training set exists. Trained on this data set, the network densely segments new volumetric images. The proposed network extends the previous u-net architecture from Ronneberger et al. by replacing all 2D operations with their 3D counterparts. The implementation performs on-the-fly elastic deformations for efficient data augmentation during training. It is trained end-to-end from scratch, i.e., no pre-trained network is required. We test the performance of the proposed method on a complex, highly variable 3D structure, the Xenopus kidney, and achieve good results for both use cases.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

A Multi-cut Formulation for Joint Segmentation and Tracking of Multiple Objects

Recently, Minimum Cost Multicut Formulations have been proposed and proven to be successful in both motion trajectory segmentation and multi-target tracking scenarios. Both tasks benefit from decomposing a graphical model into an optimal number of connected components based on attractive and repulsive pairwise terms. The two tasks are formulated on different levels of granularity and, accordingly, leverage mostly local information for motion segmentation and mostly high-level information for multi-target tracking. In this paper we argue that point trajectories and their local relationships can contribute to the high-level task of multi-target tracking and also argue that high-level cues from object detection and tracking are helpful to solve motion segmentation. We propose a joint graphical model for point trajectories and object detections whose Multicuts are solutions to motion segmentation {\it and} multi-target tracking problems at once. Results on the FBMS59 motion segmentation benchmark as well as on pedestrian tracking sequences from the 2D MOT 2015 benchmark demonstrate the promise of this joint approach.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Artistic style transfer for videos

In the past, manually re-drawing an image in a certain artistic style required a professional artist and a long time. Doing this for a video sequence single-handed was beyond imagination. Nowadays computers provide new possibilities. We present an approach that transfers the style from one image (for example, a painting) to a whole video sequence. We make use of recent advances in style transfer in still images and propose new initializations and loss functions applicable to videos. This allows us to generate consistent and stable stylized video sequences, even in cases with large motion and strong occlusion. We show that the proposed method clearly outperforms simpler baselines both qualitatively and quantitatively.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networks

The FlowNet demonstrated that optical flow estimation can be cast as a learning problem. However, the state of the art with regard to the quality of the flow has still been defined by traditional methods. Particularly on small displacements and real-world data, FlowNet cannot compete with variational methods. In this paper, we advance the concept of end-to-end learning of optical flow and make it work really well. The large improvements in quality and speed are caused by three major contributions: first, we focus on the training data and show that the schedule of presenting data during training is very important. Second, we develop a stacked architecture that includes warping of the second image with intermediate optical flow. Third, we elaborate on small displacements by introducing a sub-network specializing on small motions. FlowNet 2.0 is only marginally slower than the original FlowNet but decreases the estimation error by more than 50%. It performs on par with state-of-the-art methods, while running at interactive frame rates. Moreover, we present faster variants that allow optical flow computation at up to 140fps with accuracy matching the original FlowNet.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Generating Images with Perceptual Similarity Metrics based on Deep Networks

Image-generating machine learning models are typically trained with loss functions based on distance in the image space. This often leads to over-smoothed results. We propose a class of loss functions, which we call deep perceptual similarity metrics (DeePSiM), that mitigate this problem. Instead of computing distances in the image space, we compute distances between image features extracted by deep neural networks. This metric better reflects perceptually similarity of images and thus leads to better results. We show three applications: autoencoder training, a modification of a variational autoencoder, and inversion of deep convolutional networks. In all cases, the generated images look sharp and resemble natural images.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Inverting Visual Representations with Convolutional Networks

Feature representations, both hand-designed and learned ones, are often hard to analyze and interpret, even when they are extracted from visual data. We propose a new approach to study image representations by inverting them with an up-convolutional neural network. We apply the method to shallow representations (HOG, SIFT, LBP), as well as to deep networks. For shallow representations our approach provides significantly better reconstructions than existing methods, revealing that there is surprisingly rich information contained in these features. Inverting a deep network trained on ImageNet provides several insights into the properties of the feature representation learned by the network. Most strikingly, the colors and the rough contours of an image can be reconstructed from activations in higher network layers and even from the predicted class probabilities.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Multi-view 3D Models from Single Images with a Convolutional Network

We present a convolutional network capable of inferring a 3D representation of a previously unseen object given a single image of this object. Concretely, the network can predict an RGB image and a depth map of the object as seen from an arbitrary view. Several of these depth maps fused together give a full point cloud of the object. The point cloud can in turn be transformed into a surface mesh. The network is trained on renderings of synthetic 3D models of cars and chairs. It successfully deals with objects on cluttered background and generates reasonable predictions for real images of cars.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Object Detection, Tracking, and Motion Segmentation for Object-level Video Segmentation

We present an approach for object segmentation in videos that combines frame-level object detection with concepts from object tracking and motion segmentation. The approach extracts temporally consistent object tubes based on an off-the-shelf detector. Besides the class label for each tube, this provides a location prior that is independent of motion. For the final video segmentation, we combine this information with motion cues. The method overcomes the typical problems of weakly supervised/unsupervised video segmentation, such as scenes with no motion, dominant camera motion, and objects that move as a unit. In contrast to most tracking methods, it provides an accurate, temporally consistent segmentation of each object. We report results on four video segmentation datasets: YouTube Objects, SegTrackv2, egoMotion, and FBMS.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Pixel-level Encoding and Depth Layering for Instance-level Semantic Labeling

Recent approaches for instance-aware semantic labeling have augmented convolutional neural networks (CNNs) with complex multi-task architectures or computationally expensive graphical models. We present a method that leverages a fully convolutional network (FCN) to predict semantic labels, depth and an instance-based encoding using each pixel's direction towards its corresponding instance center. Subsequently, we apply low-level computer vision techniques to generate state-of-the-art instance segmentation on the street scene datasets KITTI and Cityscapes. Our approach outperforms existing works by a large margin and can additionally predict absolute distances of individual instances from a monocular image as well as a pixel-level semantic labeling.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Point-wise mutual information-based video segmentation with high temporal consistency

In this paper, we tackle the problem of temporally consistent boundary detection and hierarchical segmentation in videos. While finding the best high-level reasoning of region assignments in videos is the focus of much recent research, temporal consistency in boundary detection has so far only rarely been tackled. We argue that temporally consistent boundaries are a key component to temporally consistent region assignment. The proposed method is based on the point-wise mutual information (PMI) of spatio-temporal voxels. Temporal consistency is established by an evaluation of PMI-based point affinities in the spectral domain over space and time. Thus, the proposed method is independent of any optical flow computation or previously learned motion models. The proposed low-level video segmentation method outperforms the learning-based state of the art in terms of standard region metrics.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Synthesizing the preferred inputs for neurons in neural networks via deep generator networks

Deep neural networks (DNNs) have demonstrated state-of-the-art results on many pattern recognition tasks, especially vision classification problems. Understanding the inner workings of such computational brains is both fascinating basic science that is interesting in its own right - similar to why we study the human brain - and will enable researchers to further improve DNNs. One path to understanding how a neural network functions internally is to study what each of its neurons has learned to detect. One such method is called activation maximization (AM), which synthesizes an input (e.g. an image) that highly activates a neuron. Here we dramatically improve the qualitative state of the art of activation maximization by harnessing a powerful, learned prior: a deep generator network (DGN). The algorithm (1) generates qualitatively state-of-the-art synthetic images that look almost real, (2) reveals the features learned by each neuron in an interpretable way, (3) generalizes well to new datasets and somewhat well to different network architectures without requiring the prior to be relearned, and (4) can be considered as a high-quality generative method (in this case, by generating novel, creative, interesting, recognizable images).

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Techniques for Gradient Based Bilevel Optimization with Nonsmooth Lower Level Problems

We propose techniques for approximating bilevel optimization problems with non-smooth lower level problems that can have a non-unique solution. To this end, we substitute the expression of a minimizer of the lower level minimization problem with an iterative algorithm that is guaranteed to converge to a minimizer of the problem. Using suitable non-linear proximal distance functions, the update mappings of such an iterative algorithm can be differentiable, notwithstanding the fact that the minimization problem is non-smooth.

0 citations0 reviewsNo code linkedOpen access
preprint2015arXiv

Descriptor Matching with Convolutional Neural Networks: a Comparison to SIFT

Latest results indicate that features learned via convolutional neural networks outperform previous descriptors on classification tasks by a large margin. It has been shown that these networks still work well when they are applied to datasets or recognition tasks different from those they were trained on. However, descriptors like SIFT are not only used in recognition but also for many correspondence problems that rely on descriptor matching. In this paper we compare features from various layers of convolutional neural nets to standard SIFT descriptors. We consider a network that was trained on ImageNet and another one that was trained without supervision. Surprisingly, convolutional neural networks clearly outperform SIFT on descriptor matching. This paper has been merged with arXiv:1406.6909

0 citations0 reviewsNo code linkedOpen access
preprint2015arXiv

Discriminative Unsupervised Feature Learning with Exemplar Convolutional Neural Networks

Deep convolutional networks have proven to be very successful in learning task specific features that allow for unprecedented performance on various computer vision tasks. Training of such networks follows mostly the supervised learning paradigm, where sufficiently many input-output pairs are required for training. Acquisition of large training sets is one of the key challenges, when approaching a new task. In this paper, we aim for generic feature learning and present an approach for training a convolutional network using only unlabeled data. To this end, we train the network to discriminate between a set of surrogate classes. Each surrogate class is formed by applying a variety of transformations to a randomly sampled 'seed' image patch. In contrast to supervised network training, the resulting feature representation is not class specific. It rather provides robustness to the transformations that have been applied during training. This generic feature representation allows for classification results that outperform the state of the art for unsupervised learning on several popular datasets (STL-10, CIFAR-10, Caltech-101, Caltech-256). While such generic features cannot compete with class specific features from supervised training on a classification task, we show that they are advantageous on geometric matching problems, where they also outperform the SIFT descriptor.

0 citations0 reviewsNo code linkedOpen access
preprint2015arXiv

Efficient Decomposition of Image and Mesh Graphs by Lifted Multicuts

Formulations of the Image Decomposition Problem as a Multicut Problem (MP) w.r.t. a superpixel graph have received considerable attention. In contrast, instances of the MP w.r.t. a pixel grid graph have received little attention, firstly, because the MP is NP-hard and instances w.r.t. a pixel grid graph are hard to solve in practice, and, secondly, due to the lack of long-range terms in the objective function of the MP. We propose a generalization of the MP with long-range terms (LMP). We design and implement two efficient algorithms (primal feasible heuristics) for the MP and LMP which allow us to study instances of both problems w.r.t. the pixel grid graphs of the images in the BSDS-500 benchmark. The decompositions we obtain do not differ significantly from the state of the art, suggesting that the LMP is a competitive formulation of the Image Decomposition Problem. To demonstrate the generality of the LMP, we apply it also to the Mesh Decomposition Problem posed by the Princeton benchmark, obtaining state-of-the-art decompositions.

0 citations0 reviewsNo code linkedOpen access
preprint2015arXiv

FlowNet: Learning Optical Flow with Convolutional Networks

Convolutional neural networks (CNNs) have recently been very successful in a variety of computer vision tasks, especially on those linked to recognition. Optical flow estimation has not been among the tasks where CNNs were successful. In this paper we construct appropriate CNNs which are capable of solving the optical flow estimation problem as a supervised learning task. We propose and compare two architectures: a generic architecture and another one including a layer that correlates feature vectors at different image locations. Since existing ground truth data sets are not sufficiently large to train a CNN, we generate a synthetic Flying Chairs dataset. We show that networks trained on this unrealistic data still generalize very well to existing datasets such as Sintel and KITTI, achieving competitive accuracy at frame rates of 5 to 10 fps.

0 citations0 reviewsNo code linkedOpen access
preprint2015arXiv

Striving for Simplicity: The All Convolutional Net

Most modern convolutional neural networks (CNNs) used for object recognition are built using the same principles: Alternating convolution and max-pooling layers followed by a small number of fully connected layers. We re-evaluate the state of the art for object recognition from small images with convolutional networks, questioning the necessity of different components in the pipeline. We find that max-pooling can simply be replaced by a convolutional layer with increased stride without loss in accuracy on several image recognition benchmarks. Following this finding -- and building on other recent work for finding simple network structures -- we propose a new architecture that consists solely of convolutional layers and yields competitive or state of the art performance on several object recognition datasets (CIFAR-10, CIFAR-100, ImageNet). To analyze the network we introduce a new variant of the "deconvolution approach" for visualizing features learned by CNNs, which can be applied to a broader range of network structures than existing approaches.

0 citations0 reviewsNo code linkedOpen access
preprint2015arXiv

U-Net: Convolutional Networks for Biomedical Image Segmentation

There is large consent that successful training of deep networks requires many thousand annotated training samples. In this paper, we present a network and training strategy that relies on the strong use of data augmentation to use the available annotated samples more efficiently. The architecture consists of a contracting path to capture context and a symmetric expanding path that enables precise localization. We show that such a network can be trained end-to-end from very few images and outperforms the prior best method (a sliding-window convolutional network) on the ISBI challenge for segmentation of neuronal structures in electron microscopic stacks. Using the same network trained on transmitted light microscopy images (phase contrast and DIC) we won the ISBI cell tracking challenge 2015 in these categories by a large margin. Moreover, the network is fast. Segmentation of a 512x512 image takes less than a second on a recent GPU. The full implementation (based on Caffe) and the trained networks are available at http://lmb.informatik.uni-freiburg.de/people/ronneber/u-net .

0 citations0 reviewsNo code linkedOpen access
preprint2014arXiv

iPiano: Inertial Proximal Algorithm for Non-Convex Optimization

In this paper we study an algorithm for solving a minimization problem composed of a differentiable (possibly non-convex) and a convex (possibly non-differentiable) function. The algorithm iPiano combines forward-backward splitting with an inertial force. It can be seen as a non-smooth split version of the Heavy-ball method from Polyak. A rigorous analysis of the algorithm for the proposed class of problems yields global convergence of the function values and the arguments. This makes the algorithm robust for usage on non-convex problems. The convergence result is obtained based on the \KL inequality. This is a very weak restriction, which was used to prove convergence for several other gradient methods. First, an abstract convergence theorem for a generic algorithm is proved, and, then iPiano is shown to satisfy the requirements of this theorem. Furthermore, a convergence rate is established for the general problem class. We demonstrate iPiano on computer vision problems: image denoising with learned priors and diffusion based image compression.

0 citations0 reviewsNo code linkedOpen access
preprint2014arXiv

Unsupervised feature learning by augmenting single images

When deep learning is applied to visual object recognition, data augmentation is often used to generate additional training data without extra labeling cost. It helps to reduce overfitting and increase the performance of the algorithm. In this paper we investigate if it is possible to use data augmentation as the main component of an unsupervised feature learning architecture. To that end we sample a set of random image patches and declare each of them to be a separate single-image surrogate class. We then extend these trivial one-element classes by applying a variety of transformations to the initial 'seed' patches. Finally we train a convolutional neural network to discriminate between these surrogate classes. The feature representation learned by the network can then be used in various vision tasks. We find that this simple feature learning algorithm is surprisingly successful, achieving competitive classification results on several popular vision datasets (STL-10, CIFAR-10, Caltech-101).

0 citations0 reviewsNo code linkedOpen access