-
Goal: Reduce model size and runtime while keeping the accuracy
TODO:
- export to ONNX: accuracy remains the same and runtime is slightly better for 2D but a bit worse for 3D images
- optimize ONNX m…
-
I really appreciate you to make such like this huge project.
I have a question that how can i use lite model with my (1024，1024）sized image?
Is there any othet way except simply interpolate it?
Tha…
-
I have tried run this model on my custom dataset and results seems to be not good or did I miss some preprocessing part?
-
When installing groundingDINO model in conda environment, we got a value error as below:
> Installing collected packages: addict, scipy, safetensors, regex, opencv-python-headless, opencv-python, f…
-
Love this work! Wanted to follow up since the repository was made roughly 8 months ago. Hopefully this will be MIT or Apache licensed.
-
Hi @xavibou , thanks for releasing the wonderful code for this.
I'm curious what versions of mmcv or mmcv-full packages you have used that works when running the training for `bash scripts/train_pr…
-
Hellod, Mohwald.
I want to conduct some comparative experiments with DSA, which is very important to me.
I tried your sample code to reproduce it, however I ran into some difficulties.
I'm going t…