Context parallel attention that accelerates DiT model inference with dynamic caching

PyTorch implementation of MAR+DiffLoss https://arxiv.org/abs/2406.11838

[Mamba-Survey-2024] Paper list for State-Space-Model/Mamba and it's Applications

[CVPR-2024] Pytorch implementation of "Misalignment-Robust Frequency Distribution Loss for Image Transformation"

Deep Network based Wassterstein distance for Image Quality Assessment

The offcial repository for DBIQA: a full-reference image quality assessment model aims at discerning appealing content deformations

Self Reproduction Code of Paper "Reducing Transformer Key-Value Cache Size with Cross-Layer Attention (MIT CSAIL)

TPAMI：Frequency-aware Feature Fusion for Dense Image Prediction

Implementation of "DiffSensei: Bridging Multi-Modal LLMs and Diffusion Models for Customized Manga Generation"

[ARXIV'24] 3DTrajMaster: Mastering 3D Trajectory for Multi-Entity Motion in Video Generation

HRDecoder: High-Resolution Decoder Network for Fundus Image Lesion Segmentation

The source code of IEEE TPAMI 2025 "Hyper-YOLO: When Visual Object Detection Meets Hypergraph Computation".

Official Implementation of OneNet

The paper collections for the autoregressive models in vision.

SealAI's stable diffusion implementation

ICLR 2024 (Spotlight) - SEAL: A Framework for Systematic Evaluation of Real-World Super-Resolution

You See it, You Got it: Learning 3D Creation on Pose-Free Videos at Scale

[ARXIV'24] StyleMaster: Stylize Your Video with Artistic Generation and Translation

[RSS 2023] Diffusion Policy Visuomotor Policy Learning via Action Diffusion

[IEEE TCSVT 2024]Devil in Shadow: Attacking NIR-VIS Heterogeneous Face Recognition via Adversarial Shadow

This is the official PyTorch implementation of "ZipAR: Accelerating Auto-regressive Image Generation through Spatial Locality"

[arXiv'24] Align3R: Aligned Monocular Depth Estimation for Dynamic Videos

🚀🚀 「大模型」3小时完全从0训练26M的小参数GPT！🌏 Train a 26M-parameter GPT from scratch in just 3 hours!

[CVPR 2024] SinSR: Diffusion-Based Image Super-Resolution in a Single Step