In pixels we trust: From Pixel Labeling to Object Localization and Scene Categorization
- 1 October 2018
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 355-361
Abstract
While there has been significant progress in solving the problems of image pixel labeling, object detection and scene classification, existing approaches normally address them separately. In this paper, we propose to tackle these problems from a bottom-up perspective, where we simply need a semantic segmentation of the scene as input. We employ the DeepLab architecture, based on the ResNet deep network, which leverages multi-scale inputs to later fuse their responses to perform a precise pixel labeling of the scene. This semantic segmentation mask is used to localize the objects and to recognize the scene, following two simple yet effective strategies. We evaluate the benefits of our solutions, performing a thorough experimental evaluation on the NYU Depth V2 dataset. Our approach achieves a performance that beats the leading results by a significant margin, defining the new state of the art in this benchmark for the three tasks comprising the scene understanding: semantic segmentation, object detection and scene categorization.Keywords
All Related Versions
This publication has 17 references indexed in Scilit:
- YOLO9000: Better, Faster, Stronger2017
- Pyramid Scene Parsing Network2017
- Exploring Context with Deep Structured Models for Semantic SegmentationIEEE Transactions on Pattern Analysis and Machine Intelligence, 2017
- Deep Residual Learning for Image Recognition2016
- Modality and Component Aware Feature Fusion for RGB-D Scene Classification2016
- Fully convolutional networks for semantic segmentation2015
- Indoor Scene Understanding with RGB-D Images: Bottom-up Segmentation, Object Detection and Semantic SegmentationInternational Journal of Computer Vision, 2014
- Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation2014
- Perceptual Organization and Recognition of Indoor Scenes from RGB-D Images2013
- Efficient Additive Kernels via Explicit Feature MapsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2012