Pixel Invisibility: Detecting Objects Invisible in Color Images

TL;DR

This paper introduces a novel algorithm that predicts pixel-level invisibility maps in color images under various lighting conditions, enhancing safety in critical applications by identifying potentially invisible objects without manual labeling.

Contribution

It proposes a new cross-modal knowledge distillation method from color to infra-red images to generate pixel invisibility indicators without manual annotations.

Findings

High accuracy of pixel invisibility masks

Effective use of mid-level features for invisibility detection

Improved object detection in infra-red imagery

Abstract

Despite recent success of object detectors using deep neural networks, their deployment on safety-critical applications such as self-driving cars remains questionable. This is partly due to the absence of reliable estimation for detectors' failure under operational conditions such as night, fog, dusk, dawn and glare. Such unquantifiable failures could lead to safety violations. In order to solve this problem, we created an algorithm that predicts a pixel-level invisibility map for color images that does not require manual labeling - that computes the probability that a pixel/region contains objects that are invisible in color domain, during various lighting conditions such as day, night and fog. We propose a novel use of cross modal knowledge distillation from color to infra-red domain using weakly-aligned image pairs from the day and construct indicators for the pixel-level invisibility based on the distances of their intermediate-level features. Quantitative experiments show the great performance of our pixel-level invisibility mask and also the effectiveness of distilled mid-level features on object detection in infra-red imagery.