1
Return

SIPTrack: Reliability-aware identity prediction for sparse-interval pig multi-object tracking with a new benchmark

delete2026-08-04
delete0
PRE
AI
F
Feiyue Xue
W
Wangjun Huang
J
Junkai Li
T
Tongguan Wang
W
Wei Chen
H
Hui Liu
金怀平 cover
金怀平 (Huaiping Jin)
H
Huan Wang *
Y
Ying Sha *
DOI：10.1016/j.eswa.2026.133910delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
• PigMOT benchmarks long-term pig tracking from sparse observations. • SIPTrack improves identity preservation under long and irregular gaps. • SI-Memory balances recent continuity with representative earlier evidence. • TCSF and RIPR refine ambiguous detections and stabilize identity recovery. • SIPTrack leads on PigMOT and generalizes effectively to PigTrack. Highlights
Keywords:
Multi-object tracking
Pig tracking
Precision livestock farming
Sparse-interval tracking
Spatio-temporal fusion

Journal

Expert Systems with Applications cover
Expert Systems with Applications
IF:
7.5
Papers:
2.9W
Citations:
10.2W

Organization

H
huazhong agricultural university
Scholars:
5.9K
Papers: 1.5K
Citations: 0
K
key laboratory of smart farming for agricultural animals
Scholars:
5
Papers: 2
Citations: 0
K
kunming university of science and technology
Scholars:
4.1K
Papers: 1.2K
Citations: 0
Cited Papers

Cited Papers

Citing Papers

Citing Papers