Road Scene Understanding through Computer Vision - Honda Research Institute USA

Road Scene Understanding through Computer Vision

Your application is being processed

Road Scene Understanding through Computer Vision

Job Number: P23INT-40
The goal of this internship will be to perform scene understanding in and around intersection. This would include detection of an intersection from large distances, and estimation of the lane structure at the different branches of an intersections – without using lidar and prior map information.
San Jose, CA

 

Key Responsibilities

  • ​Train deep learning models on our internal datasets for the problem, experiment with numerous different neural network designs and training strategies to improve performance.
  • Publish your work at conferences.

 

 

Minimum Qualifications

  • Proficiency with pytorch, python.
  • Experience with training perception models using deep learning
  • The candidate should be enrolled in a Masters or PhD in Computer Science or related field, or they must have finished their degree recently.

​

 

Bonus Qualifications

  • Publications in conferences such as CVPR, ICCV, ECCV, WACV, etc.
  • Prior experience in road scene understanding
  • Experience with vision-language models, and experience with unsupervised /weakly learning.

 

Desired Start Date 01/06/2025 

 

Duration

 

3 Months

 

Alternate Way to Apply

Send an e-mail to careers@honda-ri.com with the following:
- Subject line including the job number(s) you are applying for 
- Recent CV 
- A cover letter highlighting relevant background (Optional)

Please, do not contact our office to inquiry about your application status.