Night-time Scene Parsing with a Large Real Dataset

Abstract

Although huge progress has been made on scene analysis in recent years, most existing works assume the input images to be in day-time with good lighting conditions. In this work, we aim to address the night-time scene parsing (NTSP) problem, which has two main challenges: 1) labeled night-time data are scarce, and 2) over- and under-exposures may co-occur in the input night-time images and are not explicitly modeled in existing pipelines. To tackle the scarcity of night-time data, we collect a novel labeled dataset, named {\it NightCity}, of 4,297 real night-time images with ground truth pixel-level semantic annotations. To our knowledge, NightCity is the largest dataset for NTSP. In addition, we also propose an exposure-aware framework to address the NTSP problem through augmenting the segmentation process with explicitly learned exposure features. Extensive experiments show that training on NightCity can significantly improve NTSP performances and that our exposure-aware model outperforms the state-of-the-art methods, yielding top performances on our dataset as well as existing datasets.

13 pages, 11 figures. This paper is accepted by IEEE Transactions on Image Processing. The dataset can be accessed via https://dmcv.sjtu.edu.cn/people/phd/tanxin/NightCity/index.html

References in corpus (4)

• Weakly Supervised Adversarial Domain Adaptation for Semantic Segmentation in Urban Scenes
• A Joint Convolutional Neural Networks and Context Transfer for Street Scenes Labeling
• Double Similarity Distillation for Semantic Image Segmentation
• Gated Path Selection Network for Semantic Segmentation

Cited by in corpus (9)

• DMT: Dynamic Mutual Training for Semi-Supervised Learning
• Context-Aware Mixup for Domain Adaptive Semantic Segmentation
• Instance Segmentation in the Dark
• Boosting Night-time Scene Parsing with Learnable Frequency
• Large-Field Contextual Feature Learning for Glass Detection
• A review of advancements in low-light image enhancement using deep learning
• PIG: Prompt Images Guidance for Night-Time Scene Parsing
• Event-guided Low-light Video Semantic Segmentation
• Multi-Scale Denoising in the Feature Space for Low-Light Instance Segmentation