Post-processing for ExecuTorch vision models: coordinate mapping, detection decode, segmentation, depth, embeddings - View it on GitHub