few-shot-object-detection
few-shot-object-detection copied to clipboard
ucbdrive
Implementations of few-shot object detection benchmarks
) Traceback (most recent call last): File "tools/train_net.py", line 124, in launch( File "/home/sjk/mrch/FSCE/fsdet/engine/launch.py", line 52, in launch main_func(*args) File "tools/train_net.py", line 100, in main cfg = setup(args) File "tools/train_net.py",...
Hi， Thanks for your interesting work. The training procedure of the TFA generally includes 3 steps on MS-COCO and LVIS datasets as follows: 1. train base model with base class...
Hi there, i was wondering about the [novel weights training step ](https://github.com/ucbdrive/few-shot-object-detection/blob/master/docs/TRAIN_INST.md)of the last layer specifically for the LVIS dataset. How does this justify to be a few-shot setting when...
I think the judgment condition of **"if s not in c_data"** is to prevent the image path already contained in **box_(k-1) shot_category_train.txt** from appearing in **box_kshot_category_train.txt** If you operate according...
pretrain model cannot download ? Can someone provide the basetrain model for coco?
There are two classes in my dataset. I modified the number of classes in the yaml file of base.yaml to 1, the number of classes in the novel.yaml to 1,...
你好，我在第二阶段微调的时候，发现novelap为0，base ap正常。使用的参数为faster_rcnn_R_101_FPN_ft_all1_5shot.yaml 奇怪的是 ROI_HEADS: NUM_CLASSES: 10 OUTPUT_LAYER: "FastRCNNOutputLayers" FREEZE_FEAT: True BACKBONE: FREEZE: True PROPOSAL_GENERATOR: FREEZE: True 中我只要把backbone的freeze设置为false则训练ap不是0，一旦为true则novel ap=0，RoiHead和PROPOSAL_GENERATOR设置成True或False都不会变成0，只有backbone会这样。 求解！谢谢！
mAP
Hi, Thanks for sharing your Code. The evaluation metric in your method is mAP. We know that the AP must be between 0 and 1. Considering that the result of...
Thank you very much for your excellent work, I want to know how many kinds of data augmentation? Thank you very much!
