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：Parallel Inversion of Neural Radiance Fields for Robust Pose Estimation
论文：https://pnerfp.github.io/
提出了一种基于快速神经辐射场（NeRF）的并行优化方法，用于估计6-DoF目标姿态。给定目标的单个观察到的RGB图像，我们可以通过最小化从fastNeRF模型渲染的像素与观察到的图像中的像素之间的残差来预测相机的平移和旋转。我们将基于动量的相机外部优化程序集成到Instant Neural Graphics Primitives中，这是最近一个非常快速的NeRF实现。通过在姿态估计任务中引入并行蒙特卡罗采样，我们的方法克服了局部极小值，并在更广泛的搜索空间中提高了效率。我们还说明了采用更稳健的基于像素的损失函数来减少误差的重要性。实验证明，我们的方法可以在合成和真实基准上实现更好的泛化和鲁棒性。
最新论文整理
ECCV2022
Updated on : 21 Oct 2022
total number : 5
PalGAN: Image Colorization with Palette Generative Adversarial Networks
- 论文/Paper: http://arxiv.org/pdf/2210.11204
- 代码/Code: None
General Image Descriptors for Open World Image Retrieval using ViT CLIP
- 论文/Paper: http://arxiv.org/pdf/2210.11141
- 代码/Code: None
Visual-Semantic Contrastive Alignment for Few-Shot Image Classification
- 论文/Paper: http://arxiv.org/pdf/2210.11000
- 代码/Code: None
VTC: Improving Video-Text Retrieval with User Comments
- 论文/Paper: http://arxiv.org/pdf/2210.10820
- 代码/Code: None
Reversed Image Signal Processing and RAW Reconstruction. AIM 2022 Challenge Report
- 论文/Paper: http://arxiv.org/pdf/2210.11153
- 代码/Code: None
CVPR2022
NeurIPS
Updated on : 21 Oct 2022
total number : 11
Self-Supervised Learning via Maximum Entropy Coding
- 论文/Paper: http://arxiv.org/pdf/2210.11464
- 代码/Code: https://github.com/xinliu20/MEC.
Breaking Bad: A Dataset for Geometric Fracture and Reassembly
- 论文/Paper: http://arxiv.org/pdf/2210.11463
- 代码/Code: https://github.com/Wuziyi616/multi_part_assembly
TANGO: Text-driven Photorealistic and Robust 3D Stylization via Lighting Decomposition
- 论文/Paper: http://arxiv.org/pdf/2210.11277
- 代码/Code: None
Coordinates Are NOT Lonely -- Codebook Prior Helps Implicit Neural 3D Representations
- 论文/Paper: http://arxiv.org/pdf/2210.11170
- 代码/Code: None
PointTAD: Multi-Label Temporal Action Detection with Learnable Query Points
- 论文/Paper: http://arxiv.org/pdf/2210.11035
- 代码/Code: https://github.com/MCG-NJU/PointTAD.
Towards Sustainable Self-supervised Learning
- 论文/Paper: http://arxiv.org/pdf/2210.11016
- 代码/Code: None
Prophet Attention: Predicting Attention with Future Attention for Improved Image Captioning
- 论文/Paper: http://arxiv.org/pdf/2210.10914
- 代码/Code: None
Cluster and Aggregate: Face Recognition with Large Probe Set
- 论文/Paper: http://arxiv.org/pdf/2210.10864
- 代码/Code: https://github.com/mk-minchul/caface
Grounded Video Situation Recognition
- 论文/Paper: http://arxiv.org/pdf/2210.10828
- 代码/Code: None
On Feature Learning in the Presence of Spurious Correlations
- 论文/Paper: http://arxiv.org/pdf/2210.11369
- 代码/Code: https://github.com/izmailovpavel/spurious_feature_learning
Hierarchical classification at multiple operating points
- 论文/Paper: http://arxiv.org/pdf/2210.10929
- 代码/Code: https://github.com/jvlmdr/hiercls.
