Adaptive Affinity Fields for Semantic Segmentation
This addresses the problem of improving semantic segmentation accuracy and efficiency for computer vision applications, offering a simpler alternative to existing methods like CRFs and GANs.
The paper tackles semantic segmentation by proposing Adaptive Affinity Fields (AAF) to capture semantic relations between neighboring pixels, achieving above-par performance and robust generalization across datasets like PASCAL VOC 2012, Cityscapes, and GTA5.
Semantic segmentation has made much progress with increasingly powerful pixel-wise classifiers and incorporating structural priors via Conditional Random Fields (CRF) or Generative Adversarial Networks (GAN). We propose a simpler alternative that learns to verify the spatial structure of segmentation during training only. Unlike existing approaches that enforce semantic labels on individual pixels and match labels between neighbouring pixels, we propose the concept of Adaptive Affinity Fields (AAF) to capture and match the semantic relations between neighbouring pixels in the label space. We use adversarial learning to select the optimal affinity field size for each semantic category. It is formulated as a minimax problem, optimizing our segmentation neural network in a best worst-case learning scenario. AAF is versatile for representing structures as a collection of pixel-centric relations, easier to train than GAN and more efficient than CRF without run-time inference. Our extensive evaluations on PASCAL VOC 2012, Cityscapes, and GTA5 datasets demonstrate its above-par segmentation performance and robust generalization across domains.