Update CVPR2021.md

CVPR2021.md

@@ -3208,6 +3208,18 @@ Densely Coded Labels (DCL)是 Circular Smooth Label (CSL)的优化版本。DCL
 
 # 4. CVPR2021 Workshop
 
+[78] Good Practices and A Strong Baseline for Traffic Anomaly Detection(【AI CITY第一名】良好做法和强大的交通异常检测基准)<br>
+[paper](https://arxiv.org/abs/2105.03827) | [code](https://github.com/Endeavour10020/AICity2021-Anomaly-Detection)<br>
+
+[77] Dynamic-OFA: Runtime DNN Architecture Switching for Performance Scaling on Heterogeneous Embedded Platforms(Dynamic-OFA：用于在异构嵌入式平台上进行性能扩展的运行时DNN架构切换)<br>
+[paper](https://arxiv.org/abs/2105.03596)<br><br>
+
+[76] The iWildCam 2021 Competition Dataset(iWildCam 2021竞赛数据集)<br>
+[paper](https://arxiv.org/abs/2105.03494)<br><br>
+
+[75] Pareto-Optimal Quantized ResNet Is Mostly 4-bit(帕累托最优量化ResNet主要为4位)<br>
+[paper](https://arxiv.org/abs/2105.03536) | [code](https://github.com/google-research/google-research/tree/master/aqt)<br><br>
+
 [74] Neural 3D Scene Compression via Model Compression(通过模型压缩进行神经3D场景压缩)<br>
 [paper](https://arxiv.org/abs/2105.03120)<br><br>