The Role of Context for Object Detection and Semantic Segmentation in the Wild

By Roozbeh Mottaghi, Xianjie Chen, Xiaobai Liu, Nam-Gyu Cho, Seong-Whan Lee, Sanja Fidler, Raquel Urtasun, Alan L. Yuille. 2014

1,654 citations recorded by the source corpus.

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

Citations in the corpus

Object Detection via a Multi-region and Semantic Segmentation-Aware CNN Model (2015, neutral): “The motivation for this comes by the close connection between segmentation and detection and by the fact that segmentation related cues are empirically known to often help object detection [5, 11, 23].”
Fully convolutional networks for semantic segmentation (2014, methodology): “PASCAL-Context [29] provides whole scene annotations of PASCAL VOC 2010.”
Relation Networks for Object Detection (2017, background): “Object relation in post-processing Most early works use object relations as a post-processing step [12, 17, 46, 47, 36, 17].”
Unified Perceptual Parsing for Scene Understanding (2018, background): “object 335 ADE [2], Pascal-Context[27] mIoU & pixel acc.”
Semantic Understanding of Scenes Through the ADE20K Dataset (2016, methodology): “Such efforts include Pascal-Context [22], NYU Depth V2 [23], SUN database [34], SUN RGB-D dataset [31], CityScapes dataset [7], and OpenSurfaces [2, 3].”
UPSNet: A Unified Panoptic Segmentation Network (2019, background): “Earlier work [11, 27] focused on introducing datasets for this task and showed the importance of global context by demonstrating the gains in bayesian frameworks, whether structured or free-form.”
The Cityscapes Dataset for Semantic Urban Scene Understanding (2016, background): “The training parameters are identical to those publicly available for training on PASCAL-Context [45], except that we reduce the learning rate to account for the increased image resolution.”
COCO-Stuff: Thing and Stuff Classes in Context (2016, methodology): “PASCAL Context [38] and ADE20K [63] are the most similar datasets to COCO-Stuff.”
DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs (2016, methodology): “(2) Accuracy: we obtain state-of-art results on several challenging datasets, including the PASCAL VOC 2012 semantic segmentation benchmark [34], PASCAL-Context [35], PASCALPerson-Part [36], and Cityscapes [37].”
Generalized Decoding for Pixel, Image, and Language (2022, methodology): “[59] Roozbeh Mottaghi, Xianjie Chen, Xiaobai Liu, Nam-Gyu Cho, Seong-Whan Lee, Sanja Fidler, Raquel Urtasun, and Alan Yuille.”
Language-driven Semantic Segmentation (2022, background): “The number of class labels is dictated by the training dataset and typically ranges from tens (Everingham et al., 2015) to hundreds (Zhou et al., 2019; Mottaghi et al., 2014) of distinct categories.”
ZeroWaste Dataset: Towards Deformable Object Segmentation in Cluttered Scenes (2021, background): “Detection and Segmentation Datasets Many datasets for image segmentation have been proposed with the goal of densely recognizing general objects and “stuff” in image scenes like street view [7, 17, 61], natural scenes [9, 20, 35, 44,65], and indoor spaces [13,19,53].”
Deep Extreme Cut: From Extreme Points to Object Segmentation (2017, methodology): “ds for mini-val. Generalization to background (stuff) categories: In order to verify the performance of DEXTR in “background” classes, we trained a model using the background labels of PASCAL Context [24] (road, sky, sidewalk, building, wall, fence, grass, ground, water, ﬂoor, ceiling, mountain, and”
Object-Proposal Evaluation Protocol is ‘Gameable’ (2015, methodology): “To look at a more ﬁne-grained per category performance of proposal methods and DMPs, we presented the plots of recall values for all 80 (20 PASCAL + 60 non-PASCAL) categories for the modiﬁed PASCAL Context dataset.”
Pyramid Scene Parsing Network (2016, background): “More recent PASCAL VOC semantic segmentation and PASCAL context datasets [8, 29] include more labels with similar context, such as chair and sofa, horse and cow, etc.”
Scene Parsing through ADE20K Dataset (2017, methodology): “However, Pascal-Context still contains scenes primarily focused on 20 object classes, while SUN has noisy labels at the object level.”
Conditional Random Fields as Recurrent Neural Networks (2015, background): “Lack of such smoothness constraints can result in poor object delineation and small spurious regions in the segmentation output [31, 30, 14, 21].”
Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation (2018, background): “Models based on Fully Convolutional Networks (FCNs) [64, 49] have demonstrated significant improvement on several segmentation benchmarks [17, 52, 13, 83, 5].”