AI BENCHMARK PROFILE

DRGBT-1K

General AIMultimodal Perception

DRGBT-1K is a large-scale benchmark for dynamic RGBT tracking with 1,045 real-world sequences, 795K frame pairs, dense annotations, and a unified evaluation protocol across 20 trackers, plus an online leaderboard.

Released: 2026-07-22
Readiness: Paper only
Primary field: General AI

Why it matters

Dynamic modality and platform variations are underrepresented in tracking benchmarks; this provides systematic evaluation for robustness under real-world transitions, supporting tracker comparison.

Motivation

Dynamic RGBT (DRGBT) tracking aims to continuously localize a target when the available sensing modalities and observation platforms vary over time.

Primary resources

Paper ↗Hugging Face ↗

Benchmark Radar records only publicly supported details and links back to primary sources for verification.