🔥 News

• 2026.08: ReferTrack evaluation code and checkpoint are now released!
• 2026.07: Check out our latest work, “ReferTrack: Referring Then Tracking for Embodied Visual Tracking.” This work introduces a referring-then-tracking paradigm that first grounds a language-described target to an image-space bounding box and then decodes tracking waypoints from this decision, using TVBI tokens to preserve target motion cues over time. It achieves state-of-the-art single-view performance on EVT-Bench with robust sim-to-real transfer to legged and humanoid robots. Learn more on project page.
• 2026.05: Follow-Bench is now released.
• 2026.03: 🥳🥳 TPT-Bench is accpted by IJRR 2026.