ForeSea : AI Forensic Search with Multi-modal Queries for Video Surveillance
Hyojin Park *, Yi Li *, Janghoon Cho †, Sungha Choi † ‡, Jungsoo Lee †, Taotao Jing, Shuai Zhang, Munawar Hayat, Dashan Gao, Ning Bi, Fatih Porikli · 2026
Read ForeSea : AI Forensic Search with Multi-modal Queries for Video Surveillance by Hyojin Park *, Yi Li *, Janghoon Cho †, Sungha Choi † ‡, Jungsoo Lee †, Taotao Jing, Shuai Zhang, Munawar Hayat, Dashan Gao, Ning Bi, Fatih Porikli (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-22872.txt.
Description
Park and co-authors build ForeSea, a forensic search system for surveillance video. Searching long multi-camera footage for a target is slow and manual. Their new benchmark, ForeSeaQA, pairs footage with image-and-text queries and timed event labels. The system runs three stages: tracking filters footage, an embedding module indexes clips, and a video language model answers queries. On the benchmark, ForeSea gains on accuracy and temporal overlap over retrieval baselines. The authors call it the first benchmark of its kind for forensic search.
Catalog record
| CREATOR | Hyojin Park *, Yi Li *, Janghoon Cho †, Sungha Choi † ‡, Jungsoo Lee †, Taotao Jing, Shuai Zhang, Munawar Hayat, Dashan Gao, Ning Bi, Fatih Porikli |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 54980 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.22872