Zhuoran Zheng's Projects
A collection of resources and papers on Diffusion Models and Score-based Models, a darkhorse in the field of Generative Models
The tessellation effect is solved with the help of bilinear upsampling.
UHD Image Deblurring via Multi-scale Cubic-Mixer
Deep learning operations reinvented (for pytorch, tensorflow, jax and others)
We provide an easy way for visualizing
FD-Vision Mamba for Endoscopic Exposure Correction
Our new work in low -light (HRGNet)
Unofficial PyTorch/🤗Transformers(+Gemma) implementation of Leave No Context Behind: Efficient Infinite Context Transformers with Infini-attention
Single UHD Image Dehazing via Interpretable Pyramid Network
we propose a lightweight dual-domain network (LD2Net) for restoring UDC images.
Low-light image enhancement is a classical computer vision problem aiming to recover normal-exposure images from low-light images. However, convolutional neural networks commonly used in this field are good at sampling low-frequency local structural features in the spatial domain, which leads to unclear texture details of the reconstructed images. To alleviate this problem, we propose a novel module using the Fourier coefficients, which can recover high-quality texture details under the constraint of semantics in the frequency phase and supplement the spatial domain. In addition, we design a simple and efficient module for the image spatial domain using dilated convolutions with different receptive fields to alleviate the loss of detail caused by frequent downsampling. We integrate the above parts into an end-to-end dual branch network and design a novel loss committee and an adaptive fusion module to guide the network to flexibly combine spatial and frequency domain features to generate more pleasing visual effects. Finally, we evaluate the proposed network on public benchmarks. Extensive experimental results show that our method outperforms many existing state-of-the-art ones, showing outstanding performance and potential.
A Simple Case of Image Dehazing with the Help of NAS.
Tools for loading video dataset and transforms on video in pytorch. You can directly load video files without preprocessing.
Python-based method to apply fog and rain to images.
We use SAM to segment the required targets and then run super-resolution or deblurring.
Image restoration using spiking neural networks.
Image Restoration via UAVFormer for Under-Display Camera of UAV
Convolutional neural networks (CNNs) have achieved unparalleled success in the single Low-light Image Enhancement (LIE) task. Existing CNN-based LIE models over-focus on pixel-level reconstruction effects, hence ignoring the theoretical guidance for sustainable optimization, which hinders their application to Ultra-High Definition (UHD) images. To address the above problems, we propose a new interpretable network, which capable of performing LIE on UHD images in real time on a single GPU. The proposed network consists of two CNNs: the first part is to use the first-order unfolding Taylor’s formula to build an interpretable network, and combine two UNets in the form of first-order Taylor’s polynomials. Then we use this constructed network to extract the feature maps of the low-resolution input image, and finally process the feature maps to form a multi-dimensional tensor termed a bilateral grid that acts on the original image to yield an enhanced result. The second part is the image enhancement using the bilateral grid. In addition, we propose a polynomial channel enhancement method to enhance UHD images. Experimental results show that the proposed method significantly outperforms state-of-the-art methods for UHD LIE on a single GPU with 24G RAM (100 fps).
Ultra HD resolution multi-exposure image fusion algorithm, which employs an implicit function to generate a 3D LUT grid of arbitrary resolution to obtain a clear ultra HD image.
We create a dataset and a customized algorithm to run image super-resolution at 4K resolution in real time
[ACCV2022]
Ultra-High-Definition Image HDR Reconstruction via Collaborative Bilateral Learning.