Spaces:

---

yonigozlan

/

omdet-turbo-open-vocabulary

like 1

Runtime error

Hi, The base model (omlab/omdet-turbo-swin-tiny-hf) only supports images. How did you get this working on video?

Are you batch processing all the frames of the video? if so, can you suggest a way to run the inferencing on a live video feed?

Thanks!