Stars
[CVPR 2025] DEIM: DETR with Improved Matching for Fast Convergence
🚀🚀 「大模型」2小时完全从0训练26M的小参数GPT！🌏 Train a 26M-parameter GPT from scratch in just 2h!
Official PyTorch Implementation of "Diffusion Transformers with Representation Autoencoders"
From babyGPT to diffusion GPT: An annotated implementation of a character-level discrete diffusion model (adapted from Karpathy’s baby GPT).
Official implementation for the paper "Deep ViT Features as Dense Visual Descriptors".
[ICLR2025] Text4Seg: Reimagining Image Segmentation as Text Generation
Official Implementation of "Towards Open-Vocabulary Semantic Segmentation without Semantic Labels" (NeurIPS 2024)
A DKMS package for easy use of Realtek r8126 driver, which supports 5 GbE.
[ACL 2025 Main] InstructPart: Task-Oriented Part Segmentation with Instruction Reasoning
[AAAI 2024] TagCLIP: A Local-to-Global Framework to Enhance Open-Vocabulary Multi-Label Classification of CLIP Without Training
Reference PyTorch implementation and models for DINOv3
Collection of scripts and notebooks for OpenAI's latest GPT OSS models
A Vision-Language Model for Spatial Affordance Prediction in Robotics
LOCATE: Localize and Transfer Object Parts for Weakly Supervised Affordance Grounding (CVPR 2023)
One-Shot Open Affordance Learning with Foundation Models (CVPR 2024)
Vision Transformers Needs Registers. And Gated MLPs. And +20M params. Tiny modality gap ensues!
Official PyTorch implementation of “MaskRIS: Semantic Distortion-aware Data Augmentation for Referring Image Segmentation”
[CVPR 2025] Fine-Grained Image-Text Correspondence with Cost Aggregation for Open-Vocabulary Part Segmentation
[CVPR2023 Highlight] GRES: Generalized Referring Expression Segmentation
A benchmark dataset for GRES and GREC [CVPR2023 Highlight]
FleVRS: Towards Flexible Visual Relationship Segmentation, NeurIPS 2024
The official codebase for "CLIPGaze: Zero-Shot Goal-Directed scanpath prediction using CLIP [ICASSP 2025]"