Skip to content

This project integrates Vision-Language Models (VLMs) with autonomous driving systems to enhance decision-making through scene understanding and reasoning. Techniques like YOLOv8, SAM2, and optical flow are utilized for robust object tracking and motion estimation.

Notifications You must be signed in to change notification settings

muralikarteek7/VLM4Autonomy-

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

11 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation


FROM VISION TO UNDERSTANDING: Leveraging VLMs to enable autonomous driving decisions

  1. Optical Flow with SAM2 and YOLO-V8:

    1. Result

    2. Flowchart

  2. Optical Flow aided Ego-Vehicle Motion Estimation:

    1. Road Segmentation Mask

    2. Largest Rectangle Drivable area

    3. Optical flow of dynamic objects

    4. Optical Flow of drivable are

    5. Ego-Vehicle motion direction visualization - This shows 4 consecutive frame’s motion direction of the ego-vehicle in 3D world coordinates - estimated from 1 camera using Optical Flow


About

This project integrates Vision-Language Models (VLMs) with autonomous driving systems to enhance decision-making through scene understanding and reasoning. Techniques like YOLOv8, SAM2, and optical flow are utilized for robust object tracking and motion estimation.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •