Towards total scene understanding: Classification, annotation and segmentation in an automatic framework

By Li-Jia Li, Richard Socher, Li Fei-Fei. 2009

565 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

YFCC100M (2015, background): “Flickr, having a rich diverse collection of image types, provides the groundwork for total scene understanding [14] in computer vision and artiﬁcial intelligence, a crucial task that can be expanded upon with our dataset, and even more so once additional annotations are released by ourselves and by…”
Deep Visual-Semantic Alignments for Generating Image Descriptions (2014, neutral): “We ﬁrst present a model that aligns segments of sentences to the visual regions that they describe through a multimodal embedding.”
Every Picture Tells a Story: Generating Sentences from Images (2010, unknown): “Using a different generative model, Li and Fei-Fei demonstrate that relations improve object labels, scene labels and segmentation [9].”
Understand scene categories by objects: A semantic regularized scene classifier using Convolutional Neural Networks (2015, background): “Some previous works have demonstrated that interaction between scene and objects has the capability to promote each other [6]–[10].”
Towards Holistic Scene Understanding: Feedback Enabled Cascaded Classification Models (2010, methodology): “[4, 46] combined image classification, annotation and segmentation with a hierarchical graphical model.”
Joint People, Event, and Location Recognition in Personal Photo Collections Using Cross-Domain Context (2010, background): “In later work [16], the same authors extended this model to incorporate object segmentation.”
Harmony potentials for joint classification and segmentation (2010, methodology): “[15] employ the user tags provided by Flickr as an additional cue to infer the presence of an object in the image.”
Learning Instance Occlusion for Panoptic Segmentation (2019, methodology): “Similar methods have been developed for grammar-based models [59], scene understanding [30, 17], human pose recognition [40], segmentation [4], and object detection [28].”
Toward Holistic Scene Understanding: Feedback Enabled Cascaded Classification Models (2012, background): “Recently, several approaches have tried to combine these different classifiers for related tasks in vision [1], [2], [3], [4], [5], [6], [7], [8], [9], [10]; however, most of them tend to be ad hoc (i.e., a hard-coded rule is used) and often an intimate knowledge of the inner workings of the…”
A Tree-Based Context Model for Object Recognition (2012, background): “Bayesian hierarchical models also provide a powerful mechanism to build generative scene models [18].”
Weakly-Supervised Dual Clustering for Image Semantic Segmentation (2013, background): “Most works focus on fully or partially supervised setting which means each or partial pixels are manually labeled for model training [18, 11, 6, 22].”
Best of both worlds: Human-machine collaboration for object annotation (2015, background): “understanding [36, 63, 35] or for modeling object context [64, 14, 44, 48].”
Learning the Relative Importance of Objects from Tagged Images for Retrieval and Cross-Modal Search (2012, methodology): “Accordingly, researchers have explored a variety of ways to leverage images with associated text—including learning their correspondence for auto-annotation of regions, objects, and scenes (Duygulu et al. 2002; Barnard et al. 2003; Gupta and Davis 2008; Berg et al. 2004; Li et al. 2009; Hwang and Gr”
Individualness and Determinantal Point Processes for Pedestrian Detection (2016, background): “By learning about object locations in the image, we can better understand what is happening in the scene [2].”
Exploiting hierarchical context on a large database of object categories (2010, background): “Bayesian hierarchical models also provide a powerful mechanism to build generative scene models [15].”
Understanding web images by object relation network (2012, unknown): “The probabilistic graphical model can be a hierarchical structure [27, 14], a directed graphical model [26], or a conditional random field [10, 20, 11, 12].”