Object Specific Deep Learning Feature and Its Application to Face Detection

TL;DR

This paper introduces a method to discover and utilize object-specific features in CNNs, specifically for face detection, by fine-tuning networks to activate certain channels and creating robust heatmaps for accurate, fast detection.

Contribution

The paper presents a novel approach to identify and exploit object-specific CNN channels, particularly for face detection, enhancing detection robustness and simplicity.

Findings

Multi-resolution OSC improves face detection accuracy.

Proposed method achieves state-of-the-art face detection performance.

The approach is simple, compact, and effective in unconstrained environments.

Abstract

We present a method for discovering and exploiting object specific deep learning features and use face detection as a case study. Motivated by the observation that certain convolutional channels of a Convolutional Neural Network (CNN) exhibit object specific responses, we seek to discover and exploit the convolutional channels of a CNN in which neurons are activated by the presence of specific objects in the input image. A method for explicitly fine-tuning a pre-trained CNN to induce an object specific channel (OSC) and systematically identifying it for the human face object has been developed. Based on the basic OSC features, we introduce a multi-resolution approach to constructing robust face heatmaps for fast face detection in unconstrained settings. We show that multi-resolution OSC can be used to develop state of the art face detectors which have the advantage of being simple and compact.