MOMA: Multi-Object Multi-Actor Activity Parsing
6. prosince 2021
Řečníci
Siddharth Kapoor
Řečník · 0 sledujících
Yiyun Liang
Řečník · 0 sledujících
Michael Cooper
Řečník · 0 sledujících
Juan Carlos Niebles
Řečník · 0 sledujících
Ehsan Adeli
Řečník · 0 sledujících
Li Fei-Fei
Řečník · 6 sledujících
O prezentaci
Complex activities often involve multiple humans utilizing different objects to complete actions (e.g., in healthcare settings, physicians, nurses, and patients interact with each other and various medical devices). The recognition of these activities poses a challenge that requires a detailed understanding of actors' roles, objects' affordances, and their associated relationships. Furthermore, these purposeful activities are composed of multiple achievable steps, including sub-activities and atomic actions, which jointly define a hierarchy of action parts. This paper introduces Activity Parsing as the overarching task of temporal segmentation and classification of activities, sub-activities, atomic actions, along with an instance-level understanding of actors, objects, and their relationships in videos. Involving multiple entities (actors and objects), we argue that traditional pair-wise relationships, often used in scene or action graphs, do not appropriately represent the dynamics between them. Hence, we introduce Action Hypergraph, a spatial-temporal graph containing hyperedges (i.e., edges with higher-order relationships), as a new representation. In addition, we introduce Multi-Object Multi-Actor (MOMA), the first benchmark and dataset dedicated to activity parsing. Lastly, to parse a video, we propose the HyperGraph Activity Parsing (HGAP) network, which outperforms several baselines, including those based on regular graphs and raw video data.Complex activities often involve multiple humans utilizing different objects to complete actions (e.g., in healthcare settings, physicians, nurses, and patients interact with each other and various medical devices). The recognition of these activities poses a challenge that requires a detailed understanding of actors' roles, objects' affordances, and their associated relationships. Furthermore, these purposeful activities are composed of multiple achievable steps, including sub-activities and at…
Organizátor
NeurIPS 2021
Účet · 1,9k sledujících
O organizátorovi (NeurIPS 2021)
Neural Information Processing Systems (NeurIPS) is a multi-track machine learning and computational neuroscience conference that includes invited talks, demonstrations, symposia and oral and poster presentations of refereed papers. Following the conference, there are workshops which provide a less formal setting.
Baví vás formát? Nechte SlidesLive zachytit svou akci!
Profesionální natáčení a streamování po celém světě.
Sdílení
Doporučená videa
Prezentace na podobné téma, kategorii nebo přednášejícího
The Rigorous and Human Life Data
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Adversarial Robustness with Non-uniform Perturbations
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Two Sides of the Same Coin: Heterophily and Oversmoothing in Graph Convolutional Neural Networks
Zhlédnout později
Yujun Yan, …
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Fast Training Methods for Stochastic Compositional Optimization Problems
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Explaining machine-learned particle-flow reconstruction
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
FINE Samples for Learning with Noisy Labels
Zhlédnout později
Taehyeon Kim, …
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %