Objects in Action: An Approach for Combining Action Understanding and Object Perception

By Abhinav Kumar Gupta, Larry S. Davis. 2007

277 citations recorded by the source corpus.

Subfields: Cognitive Neuroscience, Computer Vision and Pattern Recognition, Social Psychology

How researchers see this paper

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

In Cognitive Neuroscience

In Computer Vision and Pattern Recognition

In Social Psychology

Citations in the corpus

A self-organizing neural network architecture for learning human-object interactions (2017, background): “Gupta et al. [32, 33] proposed a Bayesian network model for integrating the appearance of manipulated objects, human motion, and reactions of objects.”
Spatial-Aware Object Embeddings for Zero-Shot Localization and Classification of Actions (2017, unknown): “[14, 34, 37, 53].”
Human activity analysis (2011, background): “Similar to the above-mentioned approaches [Moore et al. 1999; Gupta and Davis 2007], their object recognition and motion estimation components were constructed to help each other.”
Learning Human-Object Interactions by Graph Parsing Neural Networks (2018, methodology): “Early work in HOI understanding studied Bayesian model [15, 16], utilized contextual relationship between human and objects [47–49], learned structured representations with spatial interaction and context [8], exploited compositional models [9], or referred to a set of HOI exemplars [19].”
Functional categorization of objects using real-time markerless motion capture (2011, background): “When the video sequences are already labeled with the object class and the motion class that are involved in object manipulation, motion and appearance cues can be combined to improve object and action recognition [20, 14, 8].”
Understanding videos, constructing plots learning a visually grounded storyline model from annotated videos (2009, background): “There has been significant research in using contextual models for action recognition [10] [20] [9] [2] [16].”
Grounded Human-Object Interaction Hotspots From Video (2018, background): “In object understanding, affordances help model object function and interaction [52, 61, 66], and have been studied jointly with hand pose/configuration [29, 53, 3] and object motion [19, 20].”
A Review of Human Activity Recognition Methods (2015, methodology): “Usually, the detection of objects in a scene may help to better understand human activities as it may provide useful information about the ongoing event (Gupta and Davis, 2007).”
A similarity measure for analyzing human activities using human-object interaction context (2014, methodology): “Table 2 demonstrates the achieved accuracies by two of the state-of-the-art algorithms (presented in [8] and [28]) on this dataset.”
Machine Recognition of Human Activities: A Survey (2008, background): “al [111] present a Bayesian network for interpreta tion of human-object interactions that integrates information fr m”
Object, Scene and Actions: Combining Multiple Features for Human Action Recognition (2010, unknown): “[12] try to improv e the localization of both objects and actions by using a graphical Bayesian model.”
Human Action Recognition With Video Data: Research and Evaluation Challenges (2014, result): “Similar to the aforementioned examples, methods developed to identify objects from human interactions with them or vice versa [25], [66], [67], [77], [78], [86], [128] also cannot be extended to actions without objects.”
Capturing Global and Local Dynamics for Human Action Recognition (2014, background): “Prediction is made by modeling the temporal and spatial relationship of sub-events [6].”
Leaving Some Stones Unturned: Dynamic Feature Prioritization for Activity Detection in Streaming Video (2016, methodology): “We chose a bag-of-objects for its strength in compactly summarizing high-level content relevant to activities [3], [4], [47].”
Real-Time RGB-D Activity Prediction by Soft Regression (2016, unknown): “ratio”, “MP”, “DCSF”, and “DO” denote observation ratio, “Moving Pose [41, 5]”, “DSTIP+DCSF [42, 5]”, and “Discriminative Order-let [5]”.”
Hollywood in Homes: Crowdsourcing Data Collection for Activity Understanding (2016, background): “Other efforts have been to collect in-house datasets for capturing human-object interactions [22] or human-human interactions [23].”
Two-Stream SR-CNNs for Action Recognition in Videos (2016, background): “This idea was deepened in the work by [7], inferring human object reaction in a graphical model, which in turn can be used to discriminate confused action classes.”
Progressively Parsing Interactional Objects for Fine Grained Action Detection (2016, methodology): “In some contextual object recognition works [17, 7, 33, 11], hands and objects are detected using different methods independently, and the detections are further fused via probabilistic graphical models for high level inference, e.”