Multi-Scale Denoising in the Feature Space for Low-Light Instance Segmentation
This addresses instance segmentation in low-light conditions, a largely unexplored domain, with incremental improvements to existing object detectors.
The paper tackles low-light instance segmentation by integrating weighted non-local blocks into the feature extractor for inherent denoising, eliminating the need for aligned ground truth during training and achieving an Average Precision improvement of at least +7.6.
Instance segmentation for low-light imagery remains largely unexplored due to the challenges imposed by such conditions, for example shot noise due to low photon count, color distortions and reduced contrast. In this paper, we propose an end-to-end solution to address this challenging task. Our proposed method implements weighted non-local blocks (wNLB) in the feature extractor. This integration enables an inherent denoising process at the feature level. As a result, our method eliminates the need for aligned ground truth images during training, thus supporting training on real-world low-light datasets. We introduce additional learnable weights at each layer in order to enhance the network's adaptability to real-world noise characteristics, which affect different feature scales in different ways. Experimental results on several object detectors show that the proposed method outperforms the pretrained networks with an Average Precision (AP) improvement of at least +7.6, with the introduction of wNLB further enhancing AP by upto +1.3.