Source author record

Jin Han

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

ResearcherUnclaimed source record

physics.optics Computer Vision

Catalog footprint

What is connected

6works

2topics

4close collaborators

Actions

Connect this record

Open graph Browse works

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.

preprint2026arXiv

DiffST: Spatiotemporal-Aware Diffusion for Real-World Space-Time Video Super-Resolution

Diffusion-based models have shown strong performance in video super-resolution (VSR) and video frame interpolation (VFI). However, their role in the coupled space-time video super-resolution (STVSR) setting remains limited. Existing diffusion-based STVSR approaches suffer from two issues: (1) low inference efficiency and (2) insufficient utilization of spatiotemporal information. These limitations impede deployment. To address these issues, we introduce DiffST, an efficient spatiotemporal-aware video diffusion framework for real-world STVSR. To improve efficiency, we adapt a pre-trained diffusion model for one-step sampling and process the entire video directly rather than operating on individual frames. Furthermore, to enhance spatiotemporal information utilization, we introduce cross-frame context aggregation (CFCA) and video representation guidance (VRG). The CFCA module aggregates information across multiple keyframes to produce intermediate frames. The VRG module extracts video-level global features to guide the diffusion process. Extensive experiments show that DiffST obtains leading results on real-world STVSR tasks. It also maintains high inference efficiency, running about 17$\times$ faster than previous diffusion-based STVSR methods. Code is available at: https://github.com/zhengchen1999/DiffST.

preprint2020arXiv

Learning Differential Diagnosis of Skin Conditions with Co-occurrence Supervision using Graph Convolutional Networks

Skin conditions are reported the 4th leading cause of nonfatal disease burden worldwide. However, given the colossal spectrum of skin disorders defined clinically and shortage in dermatology expertise, diagnosing skin conditions in a timely and accurate manner remains a challenging task. Using computer vision technologies, a deep learning system has proven effective assisting clinicians in image diagnostics of radiology, ophthalmology and more. In this paper, we propose a deep learning system (DLS) that may predict differential diagnosis of skin conditions using clinical images. Our DLS formulates the differential diagnostics as a multi-label classification task over 80 conditions when only incomplete image labels are available. We tackle the label incompleteness problem by combining a classification network with a Graph Convolutional Network (GCN) that characterizes label co-occurrence and effectively regularizes it towards a sparse representation. Our approach is demonstrated on 136,462 clinical images and concludes that the classification accuracy greatly benefit from the Co-occurrence supervision. Our DLS achieves 93.6% top-5 accuracy on 12,378 test images and consistently outperform the baseline classification network.

preprint2013arXiv

Gaussian Beam Collimation via Transformation Media

Conventional method for Gaussian beam collimation is to expand the beam waist to achieve a smaller divergence angle. Recent transformation optics offers an unconventional path to control the electromagnetic field. In this paper we show that a Gaussian beam can be converted into a plane wave-like beam by transformation media within the scale of several wavelengths. Design details and full-wave simulation results via finite element method are provided.

preprint2011arXiv

An ultra-thin waveguide twist constructed using fish-scale metallic wires

This study theoretically and experimentally investigates the transmission properties of a metamaterial slab comprised of two layers of metallic fish-scale structure arrays and a sandwiched dielectric layer. Calculations show that the asymmetric transmission can be tuned by varying the slab thickness, due to evanescent interlayer coupling. The spatial evolution of the local field inside the structure indicates that the slab functions as a perfect polarization transformer at certain frequencies in the manner of a waveguide twist. Measured transmission spectra are in good agreement with calculated results when material dissipation is considered.

preprint2011arXiv

Subwavelength electromagnetic diode: one-way response of cascading nonlinear meta-atoms

We propose a scheme for realizing subwavelength electromagnetic diode by employing cascading nonlinear meta-atoms. One-way response is demonstrated on a microwave transmission line comprising of three metallic ring resonators acting as meta-atoms and a varactor as the nonlinear medium inclusion. Experiments show that our implementation can operate simultaneously as forward diode and backward diode at different frequencies. A transmission contrast of up to 14.7dB was achieved between forward and backward transmission. Subwavelength size of our diode should be useful for miniaturization of integrated optical nanocircuits.

preprint2010arXiv

Broadband negative refraction in stacked fishnet metamaterial

We demonstrate a scheme to utilize the stacked fishnet metamaterial for all-angle negative refraction and subwavelength imaging within a wide frequency range starting from zero frequency. The theoretical predictions are verified by the finite-difference-in-time-domain (FDTD) numerical simulations. The phenomena come from the negative evanescent coupling between the adjacent slab waveguides through the breathing air holes perforated on metal layers.