-
Hi, thank you for the work and the colab.
In the colab, the following code snippet shows how to extract visual features.
```
def extract_visual_feature(video_path, ckpt_path, user_dir, is_finetune_…
-
Nice Job!
We can not find the extracted features for Flickr30k dataset from the link you provided, could you provide it?
Thanks!
-
Thank you for your interesting work!
The input size of the EgoVLP checkpoint you provided in Google drive is 16 frames, while you mentioned the size of the input video clip for EgoVLP feature is 3…
-
**The problem**:
When I tried to apply `extract_features`, it produced duplicated columns.
My script is like below,
```python:
extract_features(timeseries, column_id="id", column_sort="tim…
-
File "tracking/run_tracker.py", line 411, in
result, result_bb, fps = run_vital(img_list, init_bbox, gt=gt, savefig_dir=savefig_dir, display=display)
File "tracking/run_tracker.py", line 230,…
-
hello,
I was trying to retrain the conST model without using the trained weights conST_151673.pth, but i I encountered difficulties while performing the following step . So can you share the code ab…
-
- [x] check why automatic test/build fail
- [x] Merge Nathan's PR
-
related to https://github.com/tobast/ORandForest/issues/5
-
Step 2 in processing the features
-
Hello, your job is very good. May I ask if you would like to extract multi-scale features, flatten them, and then concatenate them? Should you concatenate them in the channel dimension or in the spati…