How to use this model for inference on videos?
#1
by dutta18 - opened
Inference with example code is given using transformers, but video inference is not provided. Please guide.
Hi @dutta18 thanks for your interest! Pls refer to https://github.com/awslabs/extending-the-context-length-of-open-source-llms/blob/main/long-llava-qwen2-7b/local_demo/long_llava_backend.py#L149 for video inference. Basically, you need to sample frames from videos and pass them as images. Thank you!