{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/joint-inference-of-groups-events-and-human","title":"Joint Inference of Groups, Events and Human Roles in Aerial Videos","arxiv_id":"1505.05957","date":"2015-05-22","proceeding":"CVPR 2015 6","authors":["Tianmin Shu","Dan Xie","Brandon Rothrock","Sinisa Todorovic","Song-Chun Zhu"],"abstract":"With the advent of drones, aerial video analysis becomes increasingly\nimportant; yet, it has received scant attention in the literature. This paper\naddresses a new problem of parsing low-resolution aerial videos of large\nspatial areas, in terms of 1) grouping, 2) recognizing events and 3) assigning\nroles to people engaged in events. We propose a novel framework aimed at\nconducting joint inference of the above tasks, as reasoning about each in\nisolation typically fails in our setting. Given noisy tracklets of people and\ndetections of large objects and scene surfaces (e.g., building, grass), we use\na spatiotemporal AND-OR graph to drive our joint inference, using Markov Chain\nMonte Carlo and dynamic programming. We also introduce a new formalism of\nspatiotemporal templates characterizing latent sub-events. For evaluation, we\nhave collected and released a new aerial videos dataset using a hex-rotor\nflying over picnic areas rich with group events. Our results demonstrate that\nwe successfully address above inference tasks under challenging conditions.","url_abs":"http://arxiv.org/abs/1505.05957v1","url_pdf":"http://arxiv.org/pdf/1505.05957v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[],"tasks":[],"methods":[],"datasets_introduced":[{"slug":"ucla-aerial-event-dataset","name":"UCLA Aerial Event Dataset","full_name":""}],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1505.05957","atlas_url":"https://app.syntology.ai/?focus=1505.05957","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}