Learning semantic relationships for better action retrieval in images

By Vignesh Ramanathan, Congcong Li, Jia Deng, Wei Han, Zhen Li, Kunlong Gu, Yang Song, Samy Bengio, Chuck Rosenberg, Li Fei-Fei. 2015

116 citations recorded by the source corpus.

Subfields: Artificial Intelligence, 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 Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Logic Tensor Networks for Semantic Image Interpretation (2017, background): “In a final group of approaches, here called language-priors, background knowledge is taken from linguistic models [Ramanathan et al., 2015; Lu et al., 2016].”
Scene Graph Generation by Iterative Message Passing (2017, background): “Recent works [6, 31] exploits more complex structures for relationship prediction.”
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.”
Weakly-Supervised Learning of Visual Relations (2017, background): “This task was studied for the special case of humanobject interactions [9, 10, 18, 39, 40, 49, 50, 51], where the triplet is in the form (person, action, object).”
Detecting and Recognizing Human-Object Interactions (2017, background): “Accurate recognition of human-object interaction can benefit numerous tasks in computer vision, such as actionspecific image retrieval [26], caption generation [35], and question answering [35, 24].”
HICO: A Benchmark for Recognizing Human-Object Interactions in Images (2015, background): “Google Image Search Ramanathan et al. [24] reported results on 27K action categories.”
Visual Translation Embedding Network for Visual Relation Detection (2017, background): “Note that there are works [3, 27, 33, 24] that exploit language priors to boost relation detection, but we are only interested in visual models.”
Relation-Aware Graph Attention Network for Visual Question Answering (2019, background): “Recent work on visual relationship [45, 43, 9] focused more on non-spatial relation, or known as “semantic”
Learning Prototypes for Visual Relationship Detection (2018, background): “The task poses a combinatorial challenge that has been tackled by exploiting statistical dependencies between the constituents of the relationships [8]–[11], leveraging semantic knowledge [10], [12]–[14] and spatial information [10], [15]–[17].”
Detecting Visual Relationships with Deep Relational Networks (2017, methodology): “In most of these studies, relationships are usually extracted using simple heuristics or hand-crafted features, and used as an auxiliary components to facilitate other tasks, such as object recognition [25–33], image classiﬁcation and retrieval [34,35], scene understanding 1 code available at…”
Vision-based human activity recognition: a survey (2020, methodology): “These methods allow the recognition of complex human activities using complex classification models, which make the specification of emotional attributes difficult [144, 228].”
A Comprehensive Survey of Scene Graphs: Generation and Application (2021, background): “In addition to drastically changing the visual appearance of the predicate, both the sparsity of the predicate representation in the training set [45], [58] and the very large predi-”
A Review of Human Activity Recognition Methods (2015, background): “Ramanathan et al. (2015) were able to transfer semantic knowledge between classes to learn human actions from still images.”
Semantic Image Interpretation - Integration of Numerical Data and Logical Knowledge for Cognitive Vision (2018, unknown): “The work in [78] predicts visual relationships by jointly considering, in the same embedding space, visual information and logical constraints (derived from WordNet) between the relationships.”