OMNIVISION Careers

Description

Position Overview:

• We are seeking an Algorithm Engineer to research, design, and implement CNN-based image signal processing (ISP) algorithms for advanced image applications. The role focuses on developing deep learning algorithms that improve image quality, enhance color fidelity, and optimize camera and sensor pipelines for real-world deployments.

Responsibilities:

1. Research, develop, and optimize CNN-based ISP algorithms, including but not limited to: demosaicing, denoising, color correction, tone mapping, defect pixel correction, and HDR reconstruction.
2. Design and implement novel image enhancement pipelines leveraging deep learning, tailored for CMOS image sensors.
3. Evaluate and benchmark algorithm performance on real datasets; iterate based on quantitative and perceptual metrics.
4. Collaborate with hardware and system teams to integrate ISP algorithms into camera pipelines or embedded systems.
5. Keep abreast of state-of-the-art research in deep learning-based ISP and propose innovative solutions for image quality enhancement.

Experience & Skills:

1. Ph.D. or M.S. (with equivalent experience) in Electrical Engineering, Computer Science, Computational Imaging, Applied Mathematics, or related fields.
2. Strong background in CNN-based ISP, low-level image processing, and computational imaging; experience with image sensor characteristics is highly desirable.
3. Extensive experience in designing and implementing algorithms independently; strong mathematical foundation preferred.
4. Proficient in C/C++ programming, with hands-on experience in Python for algorithm prototyping; familiar with deep learning frameworks.
5. Practical experience with camera ISP pipelines, color science, and perceptual image quality metrics is a plus.
6. Excellent communication skills for cross-team collaboration and technical documentation.