Describing the scene as a whole: Joint object detection, scene classification and semantic segmentation

By Jian Yao, Sanja Fidler, Raquel Urtasun. 2012

449 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Model Architectures

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Model Architectures

Citations in the corpus

Panoptic Feature Pyramid Networks (2019, background): “Panoptic segmentation: The joint task of thing and stuff segmentation has a rich history, including early work on scene parsing [51], image parsing [52], and holistic scene understanding [56].”
Panoptic Segmentation (2018, methodology): “Interestingly, while semantic and instance segmentation dominate current work, in the pre-deep learning era there was interest in the joint task described using various names such as scene parsing [42], image parsing [43], or holistic scene understanding [50].”
Learning Non-maximum Suppression (2017, methodology): “[37, 5] combine detections with semantic labelling, while [36] rephrase detection as a labelling problem.”
Pixelwise Instance Segmentation with a Dynamically Instantiated Network (2017, methodology): “This detection potential is similar to the one previously proposed by [25], [41], [45] and [48], but formulated for the differentiable mean field inference algorithm.”
The Cityscapes Dataset for Semantic Urban Scene Understanding (2016, methodology): “Following this line, [9, 37, 40, 63, 81] propose different approaches that combine the strengths of CNNs and Conditional Random Fields (CRFs).”
K-Net: Towards Unified Image Segmentation (2021, background): “There are also pioneer attempts [19, 29, 50, 62] to joint the two segmentation tasks for more comprehensive scene understanding.”
Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation (2018, methodology): “There are several model variants proposed to exploit the contextual information for segmentation [28, 65, 36, 39, 22, 79, 51, 14], including those that employ multi-scale inputs (i.”
Rethinking Atrous Convolution for Semantic Image Segmentation (2017, background): “It has been shown that global features or contextual interactions [29, 67, 38, 42, 23, 80] are beneﬁcial in correctly classifying pixels for semantic segmentation.”
Learning Deep Structured Models (2014, background): “This, however, is suboptimal as learning the non-linear classifiers ignores the dependencies between the variables of interest.”
Neuroaesthetics in fashion: Modeling the perception of fashionability (2015, background): “Most of these works follow frameworks for generic segmentation [27, 35] with additional pose-informed potentials.”
Searching for Efficient Multi-Scale Architectures for Dense Image Prediction (2018, background): “A critical element of building such systems is supplying global features and context information to perform pixel-level classification [35, 78, 41, 45, 31, 92, 60, 19, 63].”
Image Question Answering Using Convolutional Neural Network with Dynamic Parameter Prediction (2015, background): “One of the ultimate goals in computer vision is holistic scene understanding [30], which requires a system to capture various kinds of information such as objects, actions, events, scene, atmosphere, and their relations in many different levels of semantics.”
Higher Order Conditional Random Fields in Deep Neural Networks (2015, methodology): “computational costs can also be reduced if one considers pairwise interactions between superpixels rather than individual pixels [51].”
Monocular Object Instance Segmentation and Depth Ordering with CNNs (2015, background): “In [21, 40], the authors propose an energy minimization framework which jointly identifies the true positive boxes and labels the pixels with the class labels.”
"What Happens If..." Learning to Predict the Effect of Forces in Images (2016, background): “, [25,9,40]) are passive in that they are focused on predicting the scene structure, the objects, and their attributes and relations.”
Visual Relationship Detection with Language Priors (2016, background): “Some papers explicitly collected relationships in images [25,26,27,28,29] and videos [27,30,31] and helped models map these relationships from images to language.”
Learning to Fuse Things and Stuff (2018, background): “This seminal paper inspired a fertile research direction, including contributions on how to explicitly model the relations between things and stuff categories [34, 41, 37, 36, 26].”
Instance-Level Segmentation for Autonomous Driving with Deep Densely Connected MRFs (2015, methodology): “[18, 31] proposed an MRF model to identify which detections are true positive, and output pixelwise class labels, thus performing instance segmentation.”