VEC_VAD

VEC_VAD copied to clipboard

cannot find models in ./data/raw2flow

1

In the test step, the tutorial mentioned "Select the model in ./data/raw2flow, and move the files in the model folder (such as avenue_model_5raw1of_auc0.902) into ./data/raw2flow." However, I can only see...

xudanni0927

link to the pretrained object detector Cascade R-CNN

1

Thanks for your excellent work. I find the link to the pretrained object detector Cascade R-CNN in Readme.md is no longer valid when i try to download it. Could you...

naixin2410

the object detection

1

Hello, I would like to ask how you determined the detection frame of the same person in five adjacent frames in the paper “Cloze Test Helps: Effective Video Anomaly Detection...

cuixianheng

request for calc_optical_flow.py

2

Thanks for your excellent work. There are some problem when I run the calc_optical_flow.py because of the torch version missed. Could you send me or upload the UCSDpred2 optical_flow? Thank...

carolchenyx

I have tried to implement it on Google Colab but ran into some issues. Link: https://gist.github.com/mshaek/47feeae4fe500cb7b09eace9d7f3106c Can anyone share if you have implemented in google colab?

mshaek

按照说明配置的环境(mmdetection-1.0rc0)版本缺少运行（test.py）所需要的包（mmdet.datasets.pipelines）

2

朋友你好！感谢你们的工作以及对代码开源的支持！我在复现你们的工作时遇到了一个问题：在执行第4步“test on saved models”时，我发现test.py里需要用到mmdet.dataset的piplines包，但是在你们环境配置所建议的“mmdetection 1.0rc0”中，却没有piplines包，不知道这是什么原因。我检查了mmdetection的最新版本(2.0)的源码，在它的mmdet文件夹里面是有pipline包的，但是最新版本所需要的安装环境是PyTorch 1.3+，与你们工作所建议的PyTorch环境不符合。请问可否在空闲的时候，帮我解答一下这个问题？谢谢你！ 

XiaoXueHou

About mmcv question

6

Thanks for your work! When building the environment,I installed the needed package named mmcv and mmdet . However,when I run the train.py, it said "ModuleNotFoundError: No module named 'mmcv._ext'". I...

Tiacy

What is the version of the corresponding matplotlib package used? (flownet2) root@interactive14626:/opt/data/private/shh/VEC_VAD-master# python calc_optical_flow.py Traceback (most recent call last): File "calc_optical_flow.py", line 7, in from FlowNet2_src import FlowNet2 File "/opt/data/private/shh/VEC_VAD-master/FlowNet2_src/__init__.py",...

yoghurtea

The results in the paper cannot be reproduced on the avenue dataset. Can you provide detailed super parameters? Thank you!

rxp12