CV计算机视觉每日开源代码Paper with code速览-2023.11.21

点击@CV计算机视觉，关注更多CV干货

论文已打包，点击进入—>下载界面

点击加入—>CV计算机视觉交流群

1.【基础网络架构：Transformer】Multi-entity Video Transformers for Fine-Grained Video Representation Learning

• 论文地址：https://arxiv.org//pdf/2311.10873

• 开源代码：GitHub - facebookresearch/video_rep_learning: SSL Video Representation Learning project

2.【异常检测】NNG-Mix: Improving Semi-supervised Anomaly Detection with Pseudo-anomaly Generation

• 论文地址：https://arxiv.org//pdf/2311.11961

• 开源代码（即将开源）：GitHub - donghao51/NNG-Mix: Improving Semi-supervised Anomaly Detection with Pseudo-anomaly Generation

3.【语义分割】Generalized Category Discovery in Semantic Segmentation

• 论文地址：https://arxiv.org//pdf/2311.11525

• 开源代码（即将开源）：GitHub - JethroPeng/GCDSS: The official code implementation of Generalized Category Discovery in Semantic Segmentation

4.【3D目标检测】Sparse4D v3: Advancing End-to-End 3D Detection and Tracking

• 论文地址：https://arxiv.org//pdf/2311.11722

• 开源代码：GitHub - linxuewu/Sparse4D: Sparse4D v1 & v2

5.【点云】Point Cloud Self-supervised Learning via 3D to Multi-view Masked Autoencoder

• 论文地址：https://arxiv.org//pdf/2311.10887

• 开源代码（即将开源）：GitHub - Zhimin-C/Multiview-MAE

6.【点云3D目标检测】Domain Generalization of 3D Object Detection by Density-Resampling

• 论文地址：https://arxiv.org//pdf/2311.10845

• 代码即将开源

7.【医学图像分割】SA-Med2D-20M Dataset: Segment Anything in 2D Medical Imaging with 20 Million masks

• 论文地址：https://arxiv.org//pdf/2311.11969

• 开源代码：GitHub - OpenGVLab/SAM-Med2D: Official implementation of SAM-Med2D

8.【多模态】VLM-Eval: A General Evaluation on Video Large Language Models

• 论文地址：https://arxiv.org//pdf/2311.11865

• 代码即将开源

9.【多模态】LION : Empowering Multimodal Large Language Model with Dual-Level Visual Knowledge

• 论文地址：https://arxiv.org//pdf/2311.11860

• 开源代码（即将开源）：GitHub - rshaojimmy/JiuTian: JiuTian, a Multimodal Large Language Model from HITSZ

10.【多模态】CORE-MM: Complex Open-Ended Reasoning Evaluation For Multi-Modal Large Language Models

• 论文地址：https://arxiv.org//pdf/2311.11567

• 工程主页：CORE-MM: Complex Open-ended Reasoning Evaluation for Multi-modal LargeLanguage Models

• 开源代码（即将开源）：GitHub - core-mm/core-mm

11.【多模态】GPT-4V(ision) for Robotics: Multimodal Task Planning from Human Demonstration

• 论文地址：https://arxiv.org//pdf/2311.12015

• 工程主页：https://microsoft.github.io/GPT4Vision-Robot-Manipulation-Prompts/

• 代码即将开源

12.【数字人】Semantic-Preserved Point-based Human Avatar

• 论文地址：https://arxiv.org//pdf/2311.11614

• 开源代码（即将开源）：GitHub - l1346792580123/spa

13.【自动驾驶】A Language Agent for Autonomous Driving

• 论文地址：https://arxiv.org//pdf/2311.10813

• 开源代码：GitHub - USC-GVL/Agent-Driver: A Language Agent for Autonomous Driving

14.【Diffusion】Reti-Diff: Illumination Degradation Image Restoration with Retinex-based Latent Diffusion Model

• 论文地址：https://arxiv.org//pdf/2311.11638

• 开源代码（即将开源）：GitHub - ChunmingHe/Reti-Diff

15.【人体姿态估计】Multiple View Geometry Transformers for 3D Human Pose Estimation

• 论文地址：https://arxiv.org//pdf/2311.10983

• 开源代码（即将开源）：GitHub - XunshanMan/MVGFormer

16.【人群计数】Evaluating Supervision Levels Trade-Offs for Infrared-Based People Counting

• 论文地址：https://arxiv.org//pdf/2311.11974

• 开源代码（即将开源）：tortueTortue/IRPeopleCounting · GitHub

17.【图像修复】Deep Equilibrium Diffusion Restoration with Parallel Sampling

• 论文地址：https://arxiv.org//pdf/2311.11600

• 开源代码（即将开源）：GitHub - caojiezhang/DeqIR: PyTorch implementation of "Deep Equilibrium Diffusion Restoration with Parallel Sampling"

18.【NeRF】Entangled View-Epipolar Information Aggregation for Generalizable Neural Radiance Fields

• 论文地址：https://arxiv.org//pdf/2311.11845

• 开源代码（即将开源）：GitHub - tatakai1/EVENeRF

19.【三维重建】LiDAR-HMR: 3D Human Mesh Recovery from LiDAR

• 论文地址：https://arxiv.org//pdf/2311.11971

• 开源代码（即将开源）：GitHub - soullessrobot/LiDAR-HMR: Code and data for LiDAR-HMR: 3D Human Mesh Recovery from LiDAR

论文已打包，下载链接

CV计算机视觉交流群

群内包含目标检测、图像分割、目标跟踪、Transformer、多模态、NeRF、GAN、缺陷检测、显著目标检测、关键点检测、超分辨率重建、SLAM、人脸、OCR、生物医学图像、三维重建、姿态估计、自动驾驶感知、深度估计、视频理解、行为识别、图像去雾、图像去雨、图像修复、图像检索、车道线检测、点云目标检测、点云分割、图像压缩、运动预测、神经网络量化、网络部署等多个领域的大佬，不定期分享技术知识、面试技巧和内推招聘信息。

想进群的同学请添加微信号联系管理员：PingShanHai666。添加好友时请备注：学校/公司+研究方向+昵称。

推荐阅读：

CV计算机视觉每日开源代码Paper with code速览-2023.11.20

CV计算机视觉每日开源代码Paper with code速览-2023.11.17

CV计算机视觉每日开源代码Paper with code速览-2023.11.16