Update ICCV2021.md

ICCV2021.md

@@ -129,6 +129,7 @@
 * [图神经网络(GNN)](#GNN)
 * [神经网络架构搜索(NAS)](#NAS)
 * [损失函数(Loss Function)](#LossFunction)
+* [可视化/可解释性(Visualization/Interpretability)](#Visualization)
 
 ### [19. 模型压缩(Model Compression)](#ModelCompression)
 
@@ -192,6 +193,9 @@
 
 ### 2D目标检测(2D Object Detection)
 
+[11] Vector-Decomposed Disentanglement for Domain-Invariant Object Detection<br>
+[paper](https://arxiv.org/abs/2108.06685)<br><br>
+
 [10] Oriented R-CNN for Object Detection<br>
 [paper](https://arxiv.org/abs/2108.05699) | [code](https://github.com/jbwang1997/OBBDetection)<br><br>
 
@@ -229,6 +233,9 @@
 
 ### 3D目标检测(3D Object Detection)
 
+[4] Is Pseudo-Lidar needed for Monocular 3D Object detection?<br>
+[paper](https://arxiv.org/abs/2108.06417)<br><br>
+
 [3] Fog Simulation on Real LiDAR Point Clouds for 3D Object Detection in Adverse Weather<br>
 [paper](https://arxiv.org/abs/2108.05249) | [code](http://www.trace.ethz.ch/lidar_fog_simulation)<br><br>
 
@@ -299,6 +306,9 @@
 
 ### 语义分割(Semantic Segmentation)
 
+[16] Exploiting a Joint Embedding Space for Generalized Zero-Shot Semantic Segmentation<br>
+[paper](https://arxiv.org/abs/2108.06536)<br><br>
+
 [15] LabOR: Labeling Only if Required for Domain Adaptive Semantic Segmentation(Oral)<br>
 [paper](https://arxiv.org/abs/2108.05570)<br><br>
 
@@ -394,6 +404,13 @@
 [1] HeadGAN: One-shot Neural Head Synthesis and Editing<br>
 [paper](https://arxiv.org/abs/2012.08261)<br><br>
 
+<a name="FaceAnti-Spoofing"/> 
+
+### 人脸伪造/反欺骗(Face Forgery/Face Anti-Spoofing)
+
+[1] Exploring Temporal Coherence for More General Video Face Forgery Detection<br>
+[paper](https://arxiv.org/abs/2108.06747) | [code](https://github.com/easton-cau/SOTR)<br><br>
+
 
 <br>
 
@@ -415,6 +432,9 @@
 
 ### 点云(Point Cloud)
 
+[8] PICCOLO: Point Cloud-Centric Omnidirectional Localization<br>
+[paper](https://arxiv.org/abs/2108.06545)<br><br>
+
 [7] AdaFit: Rethinking Learning-based Normal Estimation on Point Clouds<br>
 [paper](https://arxiv.org/abs/2108.05836) | [code](https://runsong123.github.io/AdaFit/)<br><br>
 
@@ -561,6 +581,13 @@
 [1] Orthogonal Projection Loss<br>
 [paper](https://arxiv.org/abs/2103.14021) | [code](https://github.com/kahnchana/opl)<br><br>
 
+<a name="Visualization"/> 
+
+### 可视化/可解释性(Visualization/Interpretability)
+
+[1] Finding Representative Interpretations on Convolutional Neural Networks<br>
+[paper](https://arxiv.org/abs/2108.06384)<br><br>
+
 <br>
 
 <a name="ModelTraining"/> 
@@ -822,6 +849,9 @@
 
 ### 行为识别/行为识别/动作识别/检测/分割(Action/Activity Recognition)
 
+[6] Foreground-Action Consistency Network for Weakly Supervised Temporal Action Localization(动作定位)<br>
+[paper](https://arxiv.org/abs/2108.06524) | [code](https://github.com/LeonHLJ/FAC-Net)<br><br>
+
 [5] Learning Action Completeness from Points for Weakly-supervised Temporal Action Localization(动作定位)<br>
 [paper](https://arxiv.org/abs/2108.05029) | [code](https://github.com/Pilhyeon)<br><br>
 
@@ -864,6 +894,9 @@
 
 ## 视觉定位(Visual Localization)
 
+[4] PICCOLO: Point Cloud-Centric Omnidirectional Localization<br>
+[paper](https://arxiv.org/abs/2108.06545)<br><br>
+
 [3] Normalization Matters in Weakly Supervised Object Localization<br>
 [paper](https://arxiv.org/abs/2107.13221)<br><br>
 
@@ -1027,6 +1060,9 @@
 
 ### 表征学习(Representation Learning)
 
+[3] Collaborative Unsupervised Visual Representation Learning from Decentralized Data<br>
+[paper](https://arxiv.org/abs/2108.06492)<br><br>
+
 [2] Enhancing Self-supervised Video Representation Learning via Multi-level Feature Optimization<br>
 [paper](https://arxiv.org/abs/2108.02183)<br><br>
 
@@ -1263,6 +1299,9 @@
 
 ## 暂无分类
 
+Focus on the Positives: Self-Supervised Learning for Biodiversity Monitoring<br>
+[paper](https://arxiv.org/abs/2108.06435)<br><br>
+
 DiagViB-6: A Diagnostic Benchmark Suite for Vision Models in the Presence of Shortcut and Generalization Opportunities<br>
 [paper](https://arxiv.org/abs/2108.05779)<br><br>