Summary of Jarvis: Detecting Actions in Video Using Unified Actor-scene Context Relation Modeling, by Seok Hwan Lee et al.
JARViS: Detecting Actions in Video Using Unified Actor-Scene Context Relation Modeling
by Seok Hwan Lee, Taein Son, Soo Won Seo, Jisong Kim, Jun Won Choi
First submitted to arxiv on: 7 Aug 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed framework, Joint Actor-scene context Relation modeling based on Visual Semantics (JARViS), is a new two-stage video action detection method that leverages Transformer attention to consolidate cross-modal action semantics. This approach combines person detector-based actor features with video backbone-generated spatio-temporal scene features to model fine-grained interactions between actors and scenes. JARViS outperforms existing methods by significant margins, achieving state-of-the-art performance on three popular VAD datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary JARViS is a new way to detect actions in videos. It’s like a super smart computer vision system that looks at what people are doing in a video and what’s happening around them. JARViS uses two different types of information: where the actors are (using a special detector) and what’s going on in the scene (using a video processing system). Then, it combines this information to figure out what actions are happening in the video. This approach works really well and is better than other systems at finding actions in videos. |
Keywords
* Artificial intelligence * Attention * Semantics * Transformer
