论文/代码速递2022.10.26！

整理：AI算法与图像处理

CVPR2022论文和代码整理：https://github.com/DWCTOD/CVPR2022-Papers-with-Code-Demo

ECCV2022论文和代码整理：https://github.com/DWCTOD/ECCV2022-Papers-with-Code-Demo

最新成果demo展示：

标题：Teaching Where to Look: Attention Similarity Knowledge Distillation for Low Resolution Face Recognition

论文：https://arxiv.org/abs/2209.14498

代码：https://github.com/gist-ailab/teaching-where-to-look

深度学习在人脸识别基准测试中取得了优异的性能，但在低分辨率（LR）图像中性能显著下降。我们提出了一种注意力相似性知识提取方法，该方法将作为教师从高分辨率（HR）网络获得的注意力图转换为作为学生的LR网络，以提高LR识别性能。受人类能够基于HR图像中获得的先验知识从LR图像中近似物体区域的启发，我们利用余弦相似性设计了知识提取损失，使学生网络的注意力与教师网络的注意力相似。在各种LR人脸相关基准上的实验证实，所提出的方法总体上改善了LR设置下的识别性能，通过简单地传输构造良好的注意力图，优于最新的结果

最新论文整理

ECCV2022

Updated on : 26 Oct 2022

total number : 3

Pointly-Supervised Panoptic Segmentation

• 论文/Paper: http://arxiv.org/pdf/2210.13950
• 代码/Code: https://github.com/BraveGroup/PSPS.git.

SUPR: A Sparse Unified Part-Based Human Representation

• 论文/Paper: http://arxiv.org/pdf/2210.13861
• 代码/Code: None

Multi-Person 3D Pose and Shape Estimation via Inverse Kinematics and Refinement

• 论文/Paper: http://arxiv.org/pdf/2210.13529
• 代码/Code: None

CVPR2022

NeurIPS

Updated on : 26 Oct 2022

total number : 4

Geo-SIC: Learning Deformable Geometric Shapes in Deep Image Classifiers

• 论文/Paper: http://arxiv.org/pdf/2210.13704
• 代码/Code: https://github.com/jw4hv/Geo-SIC

Does Medical Imaging learn different Convolution Filters?

• 论文/Paper: http://arxiv.org/pdf/2210.13799
• 代码/Code: None

GLIF: A Unified Gated Leaky Integrate-and-Fire Neuron for Spiking Neural Networks

• 论文/Paper: http://arxiv.org/pdf/2210.13768
• 代码/Code: \url{https://github.com/Ikarosy/Gated-LIF}.

Learning to forecast vegetation greenness at fine resolution over Africa with ConvLSTMs

• 论文/Paper: http://arxiv.org/pdf/2210.13648
• 代码/Code: None