Unconstrained Body Recognition at Altitude and Range: Comparing Four Approaches

TL;DR

This paper compares four approaches to long-term body shape-based person identification across various challenging conditions, introducing new models and training methods with extensive datasets.

Contribution

It presents novel body identification models based on Vision Transformers and expands existing ResNet models, trained on a large, diverse dataset for robust long-term identification.

Findings

Transformers outperform ResNet models in challenging conditions

Models maintain high accuracy over large distances and altitude variations

Training on diverse datasets improves long-term identification robustness

Abstract

This study presents an investigation of four distinct approaches to long-term person identification using body shape. Unlike short-term re-identification systems that rely on temporary features (e.g., clothing), we focus on learning persistent body shape characteristics that remain stable over time. We introduce a body identification model based on a Vision Transformer (ViT) (Body Identification from Diverse Datasets, BIDDS) and on a Swin-ViT model (Swin-BIDDS). We also expand on previous approaches based on the Linguistic and Non-linguistic Core ResNet Identity Models (LCRIM and NLCRIM), but with improved training. All models are trained on a large and diverse dataset of over 1.9 million images of approximately 5k identities across 9 databases. Performance was evaluated on standard re-identification benchmark datasets (MARS, MSMT17, Outdoor Gait, DeepChange) and on an unconstrained dataset that includes images at a distance (from close-range to 1000m), at altitude (from an unmanned aerial vehicle, UAV), and with clothing change. A comparative analysis across these models provides insights into how different backbone architectures and input image sizes impact long-term body identification performance across real-world conditions.