Proceedings of the 2nd International Conference on Recent Advancement and Modernization in Sustainable Intelligent Technologies & Applications (RAMSITA-2026)

<Previous Article In Volume
Next Article In Volume>

Deep Learning Based Predictive Crowd Stampede Analysis and Rerouting

Authors
J. Manavi1, *, Maithri A. Humbarwadi1, A. Prarthana1, Prashasthi Nand Reddy1, V. Nethravathy1, K. J. Bhanushree1
1Department of Computer Science and Engineering, Bangalore Institute of Technology, Bengaluru, India
*Corresponding author. Email: manavii.jagadish@gmail.com
Corresponding Author
J. Manavi
Available Online 28 May 2026.
DOI
10.2991/978-94-6239-678-4_28How to use a DOI?
Keywords
Crowd Stampede Prediction; YOLOv8; DETR; ConvLSTM; GCN; Farneback Optical Flow; Behavioral Modeling; Rerouting
Abstract

Sudden crowd crushes at festivals, rallies, or sports events turn safe gatherings deadly in seconds when panic ripples through packed spaces. Guards watching fixed cameras can’t keep up with the chaos, missing early shoves or squeezes until it’s too late. Our setup layers fast people-spotters YOLOv8 + Deformable DETR with motion trackers ByteTrack, feeding flow maps into ConvLSTM/LSTM for speed bursts and a GCN for group pressure reads—fusing both to flag hot zones and reroute via A* paths [1-4]. Tested on MOT20’s jammed street clips, Layer 2 nails 98% motion alerts while Layer 3 hits 85% on crowd vibes, proving it catches trouble early for real fixes. Tackles blocks, group math, and live tweaks head-on.

Copyright
© 2026 The Author(s)
Open Access
Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

<Previous Article In Volume
Next Article In Volume>
Volume Title
Proceedings of the 2nd International Conference on Recent Advancement and Modernization in Sustainable Intelligent Technologies & Applications (RAMSITA-2026)
Series
Advances in Intelligent Systems Research
Publication Date
28 May 2026
ISBN
978-94-6239-678-4
ISSN
1951-6851
DOI
10.2991/978-94-6239-678-4_28How to use a DOI?
Copyright
© 2026 The Author(s)
Open Access
Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Cite this article

risenwbib
TY  - CONF
AU  - J. Manavi
AU  - Maithri A. Humbarwadi
AU  - A. Prarthana
AU  - Prashasthi Nand Reddy
AU  - V. Nethravathy
AU  - K. J. Bhanushree
PY  - 2026
DA  - 2026/05/28
TI  - Deep Learning Based Predictive Crowd Stampede Analysis and Rerouting
BT  - Proceedings of the 2nd International Conference on Recent Advancement and Modernization in Sustainable Intelligent Technologies & Applications (RAMSITA-2026)
PB  - Atlantis Press
SP  - 353
EP  - 364
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6239-678-4_28
DO  - 10.2991/978-94-6239-678-4_28
ID  - Manavi2026
ER  -