Canonical Saliency Maps: Decoding Deep Face Models
This addresses the need for trust in critical applications like law enforcement by providing interpretability for face processing models, though it is incremental as it adapts existing visualization methods to a specific domain.
The paper tackled the problem of making deep face models more transparent by introducing Canonical Saliency Maps, which project saliency onto a canonical face to highlight relevant facial areas, with results showing usefulness in understanding predictions and detecting biases.
As Deep Neural Network models for face processing tasks approach human-like performance, their deployment in critical applications such as law enforcement and access control has seen an upswing, where any failure may have far-reaching consequences. We need methods to build trust in deployed systems by making their working as transparent as possible. Existing visualization algorithms are designed for object recognition and do not give insightful results when applied to the face domain. In this work, we present 'Canonical Saliency Maps', a new method that highlights relevant facial areas by projecting saliency maps onto a canonical face model. We present two kinds of Canonical Saliency Maps: image-level maps and model-level maps. Image-level maps highlight facial features responsible for the decision made by a deep face model on a given image, thus helping to understand how a DNN made a prediction on the image. Model-level maps provide an understanding of what the entire DNN model focuses on in each task and thus can be used to detect biases in the model. Our qualitative and quantitative results show the usefulness of the proposed canonical saliency maps, which can be used on any deep face model regardless of the architecture.