2023最新CV推荐论文及代码合集

2023最新CV推荐论文及代码合集
1.【图像分割：实例分割】PatchDCT: Patch Refinement for High Quality Instance Segmentation
论文地址：https://arxiv.org/pdf/2302.02693.pdf
开源代码：https://github.com/olivia-w12/PatchDCT
2.【图像分类：Zero-Shot】CHiLS: Zero-Shot Image Classification with Hierarchical Label Sets
论文地址：https://arxiv.org/pdf/2302.02551.pdf
开源代码：https://github.com/acmi-lab/CHILS
3.【目标跟踪】MixFormer: End-to-End Tracking with Iterative Mixed Attention
论文地址：https://arxiv.org/pdf/2302.02814.pdf
开源代码：https://github.com/MCG-NJU/MixF
4.【动作识别】Fine-Grained Action Detection with RGB and Pose Information using Two Stream Convolutional Networks
论文地址：https://arxiv.org/pdf/2302.02755.pdf
开源代码：https://github.com/fidsinn/SportTaskME22
5.【动作识别】AIM: Adapting Image Models for Efficient Video Understanding
论文地址：https://arxiv.org/pdf/2302.03024.pdf
工程主页：https://adapt-image-models.github.io/
开源代码：https://github.com/taoyang1122/adapt-image-models
6.【动作识别：半监督】Pyramid Self-attention Polymerization Learning for Semi-supervised Skeleton-based Action Recognition
论文地址：https://arxiv.org/pdf/2302.02327.pdf
开源代码：https://github.com/1xbq1/PSP-Le
7.【域自适应】RLSbench: Domain Adaptation Under Relaxed Label Shift
论文地址：https://arxiv.org/pdf/2302.03020.pdf
开源代码：https://github.com/acmi-lab/RLSbench
8.【Generative Domain Adaptation】Domain Re-Modulation for Few-Shot Generative Domain Adaptation
论文地址：https://arxiv.org/pdf/2302.02550.pdf
开源代码（即将开源）：https://github.com/wuyi2020/DoR
9.【类别增量学习】Neural Collapse Inspired Feature-Classifier Alignment for Few-Shot Class-Incremental Learning
论文地址：https://arxiv.org/pdf/2302.03004.pdf
开源代码：https://github.com/NeuralCollapseApplications/FSCIL
10.【点云目标检测：室内】TR3D: Towards Real-Time Indoor 3D Object Detection
论文地址：https://arxiv.org/pdf/2302.02858.pdf
开源代码：https://github.com/SamsungLabs
11.【点云分割】Top-Down Beats Bottom-Up in 3D Instance Segmentation
论文地址：https://arxiv.org/pdf/2302.02871.pdf
开源代码：https://github.com/SamsungLabs/td3d
12.【点云：分类、分割】PaRot: Patch-Wise Rotation-Invariant Network via Feature Disentanglement and Pose Restoration
论文地址：https://arxiv.org/pdf/2302.02535.pdf
工程主页：https://patchrot.github.io/
代码即将开源
13.【自动驾驶：车道线检测】CLiNet: Joint Detection of Road Network Centerlines in 2D and 3D
论文地址：https://arxiv.org/pdf/2302.02259.pdf
开源代码：https://github.com/AutonomousVe
14.【基础网络架构：Transformer计算加速】KDEformer: Accelerating Transformers via Kernel Density Estimation
论文地址：https://arxiv.org/pdf/2302.02451.pdf
开源代码：https://github.com/majid-daliri
15.【图像去模糊】Revisiting Image Deblurring with an Efficient ConvNet
论文地址：https://arxiv.org/pdf/2302.02234.pdf
开源代码：https://github.com/lingyanruan/LaKDNet
16.【图像翻译】Zero-shot Image-to-Image Translation
论文地址：https://arxiv.org/pdf/2302.03027.pdf
开源代码（即将开源）：https://github.com/pix2pixzero/pix2pix-zero