I attended the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2025), held June 11 to 15, 2025, at Music City Center in Nashville, Tennessee. It was inspiring to meet researchers and industry leaders, exchange ideas, and see the rapid progress in computer vision, autonomous driving, robotics, and artificial intelligence.

Our team presented “Talk2Traffic: Interactive and Editable Traffic Scenario Generation for Autonomous Driving with Multimodal Large Language Model” at the CVPR 2025 Workshops. Talk2Traffic enables users to create and iteratively edit simulation-ready traffic scenarios using text, speech, and sketches.

Paper · Project page

Zilin Huang standing in front of the CVPR sign at CVPR 2025

At CVPR 2025 in Nashville.

Zilin Huang with colleagues at CVPR 2025

With colleagues at CVPR 2025.

Presenting the Talk2Traffic research poster to attendees at CVPR 2025

Presenting Talk2Traffic during the CVPR 2025 poster session.

The crowded exhibition hall at CVPR 2025

Exploring the CVPR 2025 exhibition hall.

Attendees watching a robotics demonstration at CVPR 2025

A robotics demonstration at CVPR 2025.

Attendees visiting industry exhibits at CVPR 2025

Visiting industry exhibits at CVPR 2025.

A technical session at CVPR 2025

Attending a technical session at CVPR 2025.

Downtown Nashville during CVPR 2025

Exploring downtown Nashville between conference sessions.

Broadway in Nashville illuminated at night

Nashville’s Broadway at night.