chengyu0910

Results 13 issues of chengyu0910

why do you extract 5k imgaes from val set instead of using all the val set images for novel fine-tuning?

I don't understand the 33 line in FewX-master.datasets.coco.6_voc_few_shot.py which is used to generate k-shot novel dataset final_split_voc_10_shot_instances_train2017.json file, the line 33 require that the images selected into k-shot novel dataset...

can you provide a trained model pkl file on FSOD dataset for evaluation?

Hi, thanks for your great work. When will the DLA backbone be supported?

can you provide the trained model parameters which are reported as the experiments results in you paper?

Hi, I wanna know when will the codes be released?

Hi, is the focal loss works in SSD? How much improvement is reached in VOC2007 evaluation when SSD is trained with focal loss? I want to know whether the focal...

question