yan-ctrl
yan-ctrl
I want to use your work for my own data set. Can you tell me how the annotations of boxes came about? Looking forward to your reply
> root@jy-pytorch-env-67596d5c7c-64vm2:/workspace/AlphAction-master/demo# CUDA_VISIBLE_DEVICES=5 python demo.py --video-path input001.mp4 --output-path path/data/output --cfg-path /workspace/AlphAction-master/config_files/resnet101_8x8f_denseserial.yaml --weight-path /workspace/AlphAction-master/data/models/aia_models/resnet101_8x8f_denseserial.pth Starting video demo, video path: input001.mp4 Warning: The detected frame rate 0.0 could be wrong. The behavior...
> Hi, this work is pretty cool ! I come across several problems when I use the repo. > > 1. What's the difference between ava_train_v2.2.pkl & ava_train_v2.2_with_fair_0.9.pkl ? What's...
> Is there any inference scripts on video files with out annotations? Hello, do you know how to do it now, especially how to get the files ava _ train...
> ``` > [2022-05-05 23:18:34,655][ main.py][line: 280][ INFO] Epoch [1] Iter [53880/184378] Time 0.238 (0.412) Data 0.000 (0.192) Loss 0.0788 (0.0725) > [2022-05-05 23:18:39,059][ main.py][line: 280][ INFO] Epoch [1] Iter...
> 问题已解决，单卡训练需要降低学习率，作者默认是8卡的学习率。如果是单卡跑实验，--nproc_per_node 1 这个参数要设置成1，你看看你是不是这个没改 你好，请问你在复现这篇文章时，有看过https://github.com/Siyu-C/ACAR-Net/tree/master/annotations下的[ava_train_v2.2_with_fair_0.9.pkl](https://drive.google.com/file/d/1CsCUVxdxVyZ5vUM2eGzzV42wzKxPa7bK/view?usp=sharing)文件吗，里面的'labels': [{'bounding_box': [0.093, 0.033, 0.988, 0.978], 'label': [9, 14, 58], 'person_id': [52, 52, 52]}]}的label代表的是动作标签吗，因为自己对着官方数据集的ava_train_v2,2.csv的发现坐标和person_id能对应上，但是label是不一样的，但看着又像是动作标签，现在有些不明白，希望您能回复。
对，我猜也是动作标签，是labels下的一个子内容，但确实和官方数据集ava_train_v2.2.csv的不一样，其他的都能对上，而且作者这里是不是只给了HR2O，并没有给ACFB的生成脚本。您复现的时候只用到了作者给的这几个pkl标签文件吗？ ------------------ 原始邮件 ------------------ 发件人: "Siyu-C/ACAR-Net" ***@***.***>; 发送时间: 2022年8月29日(星期一) 晚上9:23 ***@***.***>; ***@***.******@***.***>; 主题: Re: [Siyu-C/ACAR-Net] 训练过程报错 (Issue #33) 应该是动作标签 ------------------ 原始邮件 ------------------ 发件人: ***@***.***>; 发送时间: 2022年8月29日(星期一) 晚上8:53 收件人: ***@***.***>; 抄送: ***@***.***>; ***@***.***>; 主题: Re: [Siyu-C/ACAR-Net] 训练过程报错...
> Thanks for your contributions, I wanna make custom dataset(AVA) and using your net to fine tuning, but I don't know how to create AVA dataset. Could you share your...
> We are glad to provide core code here: https://gist.github.com/ShoufaChen/263eaf55599c6e884584d7fce445af45, but unfortunately don't have the bandwidth to support full code at this time. We hope you understand. > > You...