VFA
VFA copied to clipboard
csuhan
Official code of the paper "Few-Shot Object Detection via Variational Feature Aggregation" (AAAI 2023)
Hello! Thanks for your great work! I wonder whether you have the pretrained model without CAA? It would be of great help if you can provide them.
Thank you for your contribution to this amazing work. When I used the provided code for training, the map of base-training was 74.1, and the novel ap of 1-shot fine-tuning...
For example, for a 5-shot model, how should I input support set and query set?
训练时发生错误
环境：安装readme中配置的环境 系统：Linux 训练执行命令：python train.py configs/vfa/voc/vfa_split1/vfa_r101_c4_8xb4_voc-split1_base-training.py 报错：在使用voc数据集来进行base-training，开始是可以正常训练的，当训练到了3000个iter的时候，也可以成功进行checkpoint的保存，但是接着在进行验证集的推理的时候，会发生报错，如下： [ ] 0/4952, elapsed: 0s, ETA:Traceback (most recent call last): File "train.py", line 252, in main() File "train.py", line 241, in main train_detector( File...
 What should I do？
Hi,thank you for your excellent work! I only have one gpu and I have tried to modify the learning rate, but it can not achieve the accuracy you provided, I...
Hello, thank you very much for presenting such a meaningful work. I would like to ask about fig6 in the paper, how did you get the real class center? I...
Thank you for your contribution. May I ask, what is the function of PCB class and TestMixins class under utils.py file? ``` class PCB: def __init__(self, class_names, model="RN101", templates="a photo...
I used the weight file you provided for fine-tuning, but it only reached 65.1, which is far from the 67.1 you said. What are your specific parameters?
