Stars
【CVPR 2025 Highlight】MonSter: Marry Monodepth to Stereo Unleashes Power
[CVPR 2025] 3D-LLaVA: Towards Generalist 3D LMMs with Omni Superpoint Transformer
[CVPR 2025 Best Paper Nomination] FoundationStereo: Zero-Shot Stereo Matching
Cuda implementation of semi global block matching for stereo.
这是一个基于CUDA加速的快速立体匹配库，它的核心是SemiglobalMatching（SGM）算法，它不仅在时间效率上要远远优于基于CPU的常规SGM，而且占用明显更少的内存，这意味着它不仅可以在较低分辨率（百万级）图像上达到实时的帧率，且完全具备处理千万级甚至更高量级图像的能力。
CPU and CUDA implementation of Full Exhaustive Block Matching Algorithm using Integral Images
Go-ICP for globally optimal 3D pointset registration
A PyTorch implementation of PointRend: Image Segmentation as Rendering
[CVPR 2022] Learning Affinity from Attention: End-to-End Weakly-Supervised Semantic Segmentation with Transformers
Dual Attention Network for Scene Segmentation (CVPR2019)
This is the code for the NeurIPS 2019 paper Region Mutual Information Loss for Semantic Segmentation.
CCNet: Criss-Cross Attention for Semantic Segmentation (TPAMI 2020 & ICCV 2019).
Adaptive Affinity Fields for Semantic Segmentation
Pixel-Adaptive Convolutional Neural Networks (CVPR '19)
Code for the Lovász-Softmax loss 968C (CVPR 2018)
label-smooth, amsoftmax, partial-fc, focal-loss, triplet-loss, lovasz-softmax. Maybe useful
ICCV2021 (Oral) - Exploring Cross-Image Pixel Contrast for Semantic Segmentation
Learning Efficient Convolutional Networks through Network Slimming, In ICCV 2017.
Pytorch implementation of CoordConv introduced in 'An intriguing failing of convolutional neural networks and the CoordConv solution' paper. (https://arxiv.org/pdf/1807.03247.pdf)
(arXiv'23) On Point Affiliation in Feature Upsampling; the extension of (NeurIPS'22) SAPA: Similarity-Aware Point Affiliation for Feature Upsampling
(NeurIPS'22) SAPA: Similarity-Aware Point Affiliation for Feature Upsampling
This repository contains the reference implementation for our proposed Convolutional CRFs.
PyTorch implementation of "Supervised Contrastive Learning" (and SimCLR incidentally)
Google Research
[NeurIPS 2022] Geometric order learning for rank estimation