rhysdg / vision-at-a-clip

Low-latency ONNX and TensorRT based zero-shot classification and detection with contrastive language-image pre-training based prompts
16 stars 1 forks source link