Bo Qian

question about the input

2

Firstly, thank you for your excellent work! I have some doubts about inputting data. why the start index and end index in [dataset_amass_L72](https://github.com/eth-siplab/AvatarPoser/blob/db45b29d13363dcd4f836932ab776f6a2ff4ab63/data/dataset_amass.py#L72) is [frame, frame + self.window_size+1] but the...

after run the download_datasets.sh, i got an empty folder without any error log

1

i run the following instruction： `export ROBOFLOW_API_KEY = my Private API Key` the web i got my private api key is :  and then i run `bash scripts/download_datasets.sh` then...

遇到了图片没有标注导致的报错

2

作者你好，我在运行val_vp.py的时候遇到有的图片没有标注。 在这一行[preds = model.get_visual_pe(batch["img"], visual=batch["visuals"])](https://github.com/THU-MIG/yoloe/blob/ad5db223cb47fee4de9afe63b790977ae8c03559/ultralytics/models/yolo/yoloe/val.py#L34)打印出来的batch如下： {'im_file': ['/data/lvis/images/train2017/000000000072.jpg'], 'ori_shape': [[...]], 'resized_shape': [[...]], 'ratio_pad': [[...]], 'img': tensor([[[[0.4471, 0.4471, 0.4471, ..., 0.4471, 0.4471, 0.4471], [0.4471, 0.4471, 0.4471, ..., 0.4471, 0.4471, 0.4471], [0.4471, 0.4471,...

作者你好，感谢你的工作。 我在看你的代码的时候，发现训练的过程中好像没有对GT采样，而是把所有GT都作为Visual Prompts。而不像T-Rex2论文里写的，随机在所有GT中挑选N个，N的范围是1~Num_GT。 这样的训练方法会不会导致模型对visual prompt的数量不鲁棒？

Thank you very much for your work. I have some questions to ask. At the end of section 3.3, we can obtain a semantic feature of D * H *...

I performed detection on the image below using both MMGroundingDINO and GroundingDINO, with the following commands respectively:  `python demo/image_demo.py 000000002299.jpg configs/grounding_dino/grounding_dino_swin-t_pretrain_obj365_goldg_cap4m.py --weights groundingdino_swint_ogc_mmdet-822d7e9d.pth --texts 'person'` (for GroundingDINO) python demo/image_demo.py...