FeatEnHancer: Enhancing Hierarchical Features for Object Detection and Beyond Under Low-Light Vision

TL;DR

This paper introduces FeatEnHancer, a novel module that hierarchically combines multiscale features using attention guided by task loss, significantly improving low-light vision tasks like detection and segmentation.

Contribution

The work presents a general-purpose, plug-and-play module that enhances hierarchical features based on task-specific loss, outperforming prior illumination-based methods.

Findings

+5.7 mAP on ExDark object detection

+1.5 mAP on DARK FACE face detection

+5.1 mIoU on ACDC nighttime segmentation

Abstract

Extracting useful visual cues for the downstream tasks is especially challenging under low-light vision. Prior works create enhanced representations by either correlating visual quality with machine perception or designing illumination-degrading transformation methods that require pre-training on synthetic datasets. We argue that optimizing enhanced image representation pertaining to the loss of the downstream task can result in more expressive representations. Therefore, in this work, we propose a novel module, FeatEnHancer, that hierarchically combines multiscale features using multiheaded attention guided by task-related loss function to create suitable representations. Furthermore, our intra-scale enhancement improves the quality of features extracted at each scale or level, as well as combines features from different scales in a way that reflects their relative importance for the task at hand. FeatEnHancer is a general-purpose plug-and-play module and can be incorporated into any low-light vision pipeline. We show with extensive experimentation that the enhanced representation produced with FeatEnHancer significantly and consistently improves results in several low-light vision tasks, including dark object detection (+5.7 mAP on ExDark), face detection (+1.5 mAPon DARK FACE), nighttime semantic segmentation (+5.1 mIoU on ACDC ), and video object detection (+1.8 mAP on DarkVision), highlighting the effectiveness of enhancing hierarchical features under low-light vision.