mega.pytorch icon indicating copy to clipboard operation
mega.pytorch copied to clipboard

Memory Enhanced Global-Local Aggregation for Video Object Detection, CVPR2020

Results 67 mega.pytorch issues
Sort by recently updated
recently updated
newest added

你好,请问baseline是直接用faster-rcnn对每一张图片进行检测嘛? 另一个是baseline那勾选了local是什么意思呢? ![image](https://user-images.githubusercontent.com/16866386/89119040-a9367200-d4dd-11ea-83e4-ce8c9a6f4156.png)

Hello, I am trying to reproduce results for the single frame baselines with ResNet-50 and ResNet-101. For those I am using the config files `configs/vid_R_50_C4_1x.yaml` and `configs/vid_R_50_C4_1x.yaml` respectively. I get...

产生许多权重文件,如何判断那个是最好的?一个一个测试非常慢。大家有什么好的办法吗?

Can anyone share the specific network structure of MEGA?

作者您好,我是按照 README.md 上面的配置的,训练和推理也是按照那个里面的命令执行的,但是还是报错,是除了按照 README.md 里面的配资之外还要按照哪个配置啊?劳烦解答一下,谢谢!

(python38) F:\视频目标检测\代码\mega.pytorch>python -m torch.distributed.launch --nproc_per_node 4 tools/test_net.py --config-file configs/MEGA/vid_R_101_C4_MEGA_1x.yaml --motion-specific MODEL.WEIGHT MEGA_R_101.pth ***************************************** Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being...

我修改conv_body 为retinanet,但是出现错误。修改为retinanet,也就是不用RPN了,用retinanet的box,但是rpn中有version=‘’ref‘’这样的参数,retinanet中没有,这是不是意味着该codebase不支持修改底层为retinanet,只能是论文中出现的resnet系列作为底层?

Why can only one picture run on one gpu now? If I want to modify, from what code and where should I modify it? Many thanks!