Computer Vision in Low Light Scenes

at  Honda Research Institute USA

Job Number: P23INT-43
This internship concerns with developing Vision based Mapping and Localization Algorithms for Autonomous Driving with focus on low light scenes with HDR Cameras. The candidate is responsible for developing deep learning models for low light HDR image enhancement and light invariant point features. This position offers a unique opportunity to contribute to the development of next-generation mapping technologies that operate effectively in challenging lighting conditions.
San Jose, CA

MINIMUM QUALIFICATIONS

• Ph.D. or M.S student in Computer Science, Robotics, or related field.
• Familiarity with Learning based and Classical Image Enhancement and Denoising Algorithms.
• Familiarity with SOTA Vision and Deep Learning methods for estimating light invariant point features.
• Ability to transfer conceptual models and ideas to working code in Python and/or C++.
• Experience with Deep Learning frameworks PyTorch and/or TensorFlow

BONUS QUALIFICATIONS

• Hands on experience with Vision based Mapping and Localization systems.
• Publication record in top venues including CVPR, ICCV, ICRA, IROS, ECCV, NeurIPS, ICLR, etc.
• Hands on experience with Deep Learning model optimization techniques and real-time algorithms for edge devices
  Warning:You do not have the permission to access the upload fields on this form. Contact the form owner or portal administrator to request the access.

• Design and development of Deep Learning Models for low light HDR image enhancement and light invariant point features.
• Integrate these models into our large-scale Mapping & Localization system.
• Create and maintain datasets for training the network.
• Develop evaluation metrics to confirm the efficacy of proposed algorithms.