Pyramid Scene Parsing Network
arXiv (Cornell University)Published 4 December 2016Open access
Hengshuang Zhao, Jianping Shi, Xiaojuan Qi, Xiaogang Wang, Jiaya Jia
Citations312
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
Abstract
Scene parsing is challenging for unrestricted open vocabulary and diverse scenes. In this paper, we exploit the capability of global context information by different-region-based context aggregation through our pyramid pooling module together with the proposed pyramid scene parsing network (PSPNet). Our global prior representation is effective to produce good quality results on the scene parsing task, while PSPNet provides a superior framework for pixel-level prediction tasks. The proposed approach achieves state-of-the-art performance on various datasets. It came first in ImageNet scene parsing challenge 2016, PASCAL VOC 2012 benchmark and Cityscapes benchmark. A single PSPNet yields new record of mIoU accuracy 85.4% on PASCAL VOC 2012 and accuracy 80.2% on Cityscapes.
Keywords
Computer Science
arXiv (Cornell University)Very Deep Convolutional Networks for Large-Scale Image Recognition
75,540 Citations2014Karen Simonyan, Andrew Zisserman
This work investigates the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting using an architecture with very small convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers.
International Journal of Computer VisionThe Pascal Visual Object Classes (VOC) Challenge
19,539 Citations2009Mark Everingham, Luc Van Gool +3 more
arXiv (Cornell University)Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
15,635 Citations2024Sergey Ioffe
Beyond Bags of Features: Spatial Pyramid Matching for Recognizing Natural Scene Categories
7,929 Citations2006Svetlana Lazebnik, C. Schmid +1 more
This paper presents a method for recognizing scene categories based on approximate global geometric correspondence that exceeds the state of the art on the Caltech-101 database and achieves high accuracy on a large database of fifteen natural scene categories.
arXiv (Cornell University)Deep Residual Learning for Image Recognition
4,720 Citations2015Kaiming He, Xiangyu Zhang +2 more
arXiv (Cornell University)Caffe: Convolutional Architecture for Fast Feature Embedding
4,290 Citations2014Yangqing Jia, Evan Shelhamer +6 more
Lecture notes in computer scienceSpatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition
3,113 Citations2014Kaiming He, Xiangyu Zhang +2 more
This work equips the networks with another pooling strategy, “spatial pyramid pooling”, to eliminate the above requirement, and develops a new network structure, called SPP-net, which can generate a fixed-length representation regardless of image size/scale.
Semantic contours from inverse detectors
1,717 Citations2011Bharath Hariharan, Pablo Arbeláez +3 more
A simple yet effective method for combining generic object detectors with bottom-up contours to identify object contours is presented and a principled way of combining information from different part detectors and across categories is provided.
Hypercolumns for object segmentation and fine-grained localization
1,578 Citations2015Bharath Hariharan, Pablo Arbeláez +2 more
Using hypercolumns as pixel descriptors, this work defines the hypercolumn at a pixel as the vector of activations of all CNN units above that pixel, and shows results on three fine-grained localization tasks: simultaneous detection and segmentation, and keypoint localization.
arXiv (Cornell University)Multi-Scale Context Aggregation by Dilated Convolutions
1,571 Citations2015Fisher Yu, Vladlen Koltun
IEEE Transactions on Pattern Analysis and Machine IntelligenceSegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation
1,523 Citations2017Vijay Badrinarayanan, Alex Kendall +1 more
Quantitative assessments show that SegNet provides good performance with competitive inference time and most efficient inference memory-wise as compared to other architectures, including FCN and DeconvNet.
The Role of Context for Object Detection and Semantic Segmentation in the Wild
1,436 Citations2014Roozbeh Mottaghi, Xianjie Chen +6 more
A novel deformable part-based model is proposed, which exploits both local context around each candidate detection as well as global context at the level of the scene, which significantly helps in detecting objects at all scales.
arXiv (Cornell University)Going Deeper with Convolutions
1,381 Citations2014Christian Szegedy, Wei Liu +7 more
Weakly-and Semi-Supervised Learning of a Deep Convolutional Network for Semantic Image Segmentation
1,068 Citations2015George Papandreou, Liang-Chieh Chen +2 more
Expectation-Maximization (EM) methods for semantic image segmentation model training under weakly supervised and semi-supervised settings are developed and extensive experimental evaluation shows that the proposed techniques can learn models delivering competitive results on the challenging PASCAL VOC 2012 image segmentsation benchmark, while requiring significantly less annotation effort.
arXiv (Cornell University)ParseNet: Looking Wider to See Better
1,059 Citations2015Wei Liu, Andrew Rabinovich +1 more
This work presents a technique for adding global context to deep convolutional networks for semantic segmentation, and achieves state-of-the-art performance on SiftFlow and PASCAL-Context with small additional computational cost over baselines.
arXiv (Cornell University)The Cityscapes Dataset for Semantic Urban Scene Understanding
1,000 Citations2024Marius Cordts
arXiv (Cornell University)Fully Convolutional Networks for Semantic Segmentation
836 Citations2016Jonathan Long, Evan Shelhamer +1 more
arXiv (Cornell University)DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs
708 Citations2016Liang-Chieh Chen, George Papandreou +3 more
arXiv (Cornell University)Object Detectors Emerge in Deep Scene CNNs
708 Citations2014Bolei Zhou, Aditya Khosla +3 more
arXiv (Cornell University)Learning Deconvolution Network for Semantic Segmentation
614 Citations2015Hyeonwoo Noh, Seunghoon Hong +1 more
Lecture notes in computer scienceLaplacian Pyramid Reconstruction and Refinement for Semantic Segmentation
490 Citations2016Golnaz Ghiasi, Charless C. Fowlkes
A multi-resolution reconstruction architecture based on a Laplacian pyramid that uses skip connections from higher resolution feature maps and multiplicative gating to successively refine segment boundaries reconstructed from lower-resolution maps is described.
IEEE Transactions on Pattern Analysis and Machine IntelligenceNonparametric Scene Parsing via Label Transfer
386 Citations2011Ce Liu, J. Yuen +1 more
This paper proposes a novel, nonparametric approach for object recognition and scene parsing using a new technology the authors name label transfer, which is easy to implement, has few parameters, and embeds contextual information naturally in the retrieval/alignment procedure.
arXiv (Cornell University)Towards Good Practices for Very Deep Two-Stream ConvNets
385 Citations2015Limin Wang, Yuanjun Xiong +2 more
This report presents very deep two-stream ConvNets for action recognition, by adapting recent very deep architectures into video domain, and extends the Caffe toolbox into Multi-GPU implementation with high computational efficiency and low memory consumption.
arXiv (Cornell University)Scalable, High-Quality Object Detection
337 Citations2014Christian Szegedy, Scott Reed +3 more
It is demonstrated that learning-based proposal methods can effectively match the performance of hand-engineered methods while allowing for very efficient runtime-quality trade-offs.
2009 IEEE Conference on Computer Vision and Pattern RecognitionNonparametric scene parsing: Label transfer via dense scene alignment
325 Citations2009Ce Liu, Jenny Yuen +1 more
Compared to existing object recognition approaches that require training for each object category, the proposed nonparametric scene parsing system is easy to implement, has few parameters, and embeds contextual information naturally in the retrieval/alignment procedure.
arXiv (Cornell University)Semantic Understanding of Scenes through the ADE20K Dataset
189 Citations2016Bolei Zhou, Hang Zhao +5 more
arXiv (Cornell University)Semantic Image Segmentation via Deep Parsing Network
181 Citations2015Ziwei Liu, Xiaoxiao Li +3 more
Lecture notes in computer scienceZoom Better to See Clearer: Human and Object Parsing with Hierarchical Auto-Zoom Net
172 Citations2016Fangting Xia, Peng Wang +2 more
This work proposes a "Hierarchical Auto-Zoom Net" (HAZN) for object part parsing which adapts to the local scales of objects and parts and significantly outperforms the state-of-the-arts by 5% mIOU and is especially better at segmenting small instances and small parts.
arXiv (Cornell University)Bridging Category-level and Instance-level Semantic Image Segmentation
157 Citations2016Zifeng Wu, Chunhua Shen +1 more
An approach to instance-level image segmentation that is built on top of category-level segmentation, which follows a different pipeline to the popular detect-then-segment approaches that first predict instances' bounding boxes.
arXiv (Cornell University)BoxSup: Exploiting Bounding Boxes to Supervise Convolutional Networks for Semantic Segmentation
147 Citations2015Jifeng Dai, Kaiming He +1 more
arXiv (Cornell University)Attention to Scale: Scale-aware Semantic Image Segmentation
133 Citations2015Liang-Chieh Chen, Yi Yang +3 more
Are spatial and global constraints really necessary for segmentation?
71 Citations2011Aurélien Lucchi, Yunpeng Li +3 more
This investigation was unable to find evidence of a significant performance increase attributed to the introduction of spatial and consistency constraints, and found that similar levels of performance can be achieved using a much simpler design that essentially ignores these constraints.
Lecture notes in computer scienceConvolutional Scale Invariance for Semantic Segmentation
63 Citations2016Ivan Krešo, Denis Čaušević +2 more
A novel scale selection layer which extracts convolutional features at the scale which matches the corresponding reconstructed depth and frees the pixel-level classifier from the need to learn the laws of the perspective results in improved segmentation results.
arXiv (Cornell University)Feedforward semantic segmentation with zoom-out features
43 Citations2014Mohammadreza Mostajabi, Payman Yadollahpour +1 more
arXiv (Cornell University)Relay Backpropagation for Effective Learning of Deep Convolutional Neural Networks
5 Citations2015Li Shen, Zhouchen Lin +1 more