Pyramid Scene Parsing Network
By Hengshuang Zhao, Jianping Shi, Xiaojuan Qi, Xiaogang Wang, Jiaya Jia. 2016
14,412 citations recorded by the source corpus.
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. 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 the new record of mIoU accuracy 85.4% on PASCAL VOC 2012 and accuracy 80.2% on Cityscapes.
Subfields: Computer Vision and Pattern Recognition
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computer Vision and Pattern Recognition
- Luc Van Gool
- Dacheng Tao
- Ming-Hsuan Yang
- Xiaogang Wang
- Trevor Darrell
- Jitendra Malik
- Yi Yang
- Alan L. Yuille
Citations in the corpus
- Exploring the Limits of Weakly Supervised Pretraining (2018, background): “Examples include: object detection [1,4], semantic segmentation [5,6], human pose estimation [7,8], video recognition [9], monocular depth estimation [10], and so on.”
- Exploring Self-Attention for Image Recognition (2020, background): “These serve as the basis for models developed for other computer vision tasks, such as semantic segmentation [22, 3, 42, 44] and object detection [10, 9, 28, 21].”
- Axial-DeepLab: Stand-Alone Axial-Attention for Panoptic Segmentation (2020, methodology): “Some employ atrous convolutions [34,77,67,13], larger kernel [70], or image pyramids [98,85], either designed by hand or searched by algorithms [103,12,60].”
- CCNet: Criss-Cross Attention for Semantic Segmentation (2018, methodology): “Most of methods [5, 22, 40, 27, 30, 41, 35, 19, 42, 36] adopt the same backbone as ours and the others [32, 34] utilize stronger backbones.”
- Bag of Tricks for Image Classification with Convolutional Neural Networks (2018, methodology): “of other models, such as SENet [12], PSPNet [32], DeepLabV3 [1], and ShuffleNetV2 [21].”
- Unified Perceptual Parsing for Scene Understanding (2018, methodology): “[31] with a Pyramid Pooling Module (PPM) [16] appended on the last layer of the back-bone network before feeding it into the top-down branch in FPN. Top-right: We use features at various semantic levels.”
- UPSNet: A Unified Panoptic Segmentation Network (2019, methodology): “Relying on the advances in deep learning, computer vision systems have been substantially improved, especially in tasks such as semantic segmentation [40] and instance segmentation [15].”
- Panoptic Feature Pyramid Networks (2019, methodology): “To increase feature resolution, which is necessary for generating highquality results, recent top methods [12, 58, 5, 59] rely heavily on the use of dilated convolution [57] (also known as atrous convolution [10]).”
- Panoptic Segmentation (2018, methodology): “For Cityscapes, we use the val set output generated by the current leading algorithms (PSPNet [53] and Mask R-CNN [14] for semantic and instance segmentation, respectively).”
- Pixelwise Instance Segmentation with a Dynamically Instantiated Network (2017, methodology): “We use an initial semantic segmentation subnetwork that is based on the ResNet-101 architecture [52], and all of the instance unary potentials described in Sec.”
- Deep Watershed Transform for Instance Segmentation (2016, methodology): “Influence of semantic segmentation: While we elected to use PSPNet [34] as our semantic segmentation source, we additionally demonstrate that our method is able to use other sources.”
- Scaling and Benchmarking Self-Supervised Visual Representation Learning (2019, methodology): “We use the state-of-the-art PSPNet [76] architecture (implementation [78]).”
- DensePose: Dense Human Pose Estimation in the Wild (2018, methodology): “Even though we apply a multi-scale testing strategy that fuses probabilities from multiple runs using input images of different scale [49], the FCN is not sufficiently robust to deal with the variability in object scale.”
- The Reversible Residual Network: Backpropagation Without Storing Activations (2017, background): “…to the task of semantic segmentation, the performance of which is limited by a critical memory bottleneck — the input image patch needs to be large enough to process high resolution images; meanwhile, the batch size also needs to be large enough to perform effective batch normalization (e.g. [34]).”
- High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs (2017, methodology): “PSPNet [59] that we use) should be able to predict the ground truth label.”
- Multiview Transformers for Video Recognition (2022, methodology): “Inspired by ViT, ViViT [3] and Timesformer [6] were the first two works that successfully adopted a pure transformer architecture for video classification, advancing the state of the art previously set by 3D CNNs. Multiscale processing in computer vision .”
- FocalClick: Towards Practical Interactive Image Segmentation (2022, background): “The Segmentor could be any segmentation network [4, 28, 32, 41, 47] for customized scenarios.”
- EDTER: Edge Detection with Transformer (2022, background): “9 and weight decay=2e-4, and adopt a polynomial learning rate decay schedule [71] on all datasets.”