Top-down Visual Saliency Guided by Captions
This addresses the need for interpretability in captioning models for researchers and practitioners, though it is incremental as it builds on existing saliency and captioning techniques.
The paper tackles the problem of explaining region-to-word mappings in neural captioning models by proposing Caption-Guided Visual Saliency, which generates heatmaps for predicted or arbitrary sentences without pixel-level annotations. It achieves comparable captioning performance to existing methods while providing more accurate saliency heatmaps on large-scale video and image datasets.
Neural image/video captioning models can generate accurate descriptions, but their internal process of mapping regions to words is a black box and therefore difficult to explain. Top-down neural saliency methods can find important regions given a high-level semantic task such as object classification, but cannot use a natural language sentence as the top-down input for the task. In this paper, we propose Caption-Guided Visual Saliency to expose the region-to-word mapping in modern encoder-decoder networks and demonstrate that it is learned implicitly from caption training data, without any pixel-level annotations. Our approach can produce spatial or spatiotemporal heatmaps for both predicted captions, and for arbitrary query sentences. It recovers saliency without the overhead of introducing explicit attention layers, and can be used to analyze a variety of existing model architectures and improve their design. Evaluation on large-scale video and image datasets demonstrates that our approach achieves comparable captioning performance with existing methods while providing more accurate saliency heatmaps. Our code is available at visionlearninggroup.github.io/caption-guided-saliency/.