- 경험
- 어느
- 샐러리
- —
- 채용 공고
- 1
- 게시됨
- 2시간 전
- 작업 모드
- 사무실에서
- 재개하다
- 신청 시 필수 사항
당신이 일하게 될 곳
직무 설명
About the role
We’re looking for a Computer Vision Engineer to help build mirror-aware, multi-camera perception systems for live human pose tracking. The work centers on reliable 2D and 3D pose estimation, strong handling of occlusions and reflections, and efficient deployment on wearable and mobile hardware.
What you’ll do
- Develop and implement computer vision systems that work across multiple cameras and are aware of mirror reflections.
- Create pipelines for 2D and 3D human pose estimation.
- Build camera calibration, stereo vision, and multi-view reconstruction workflows.
- Detect mirrors and analyze reflections using geometric consistency methods.
- Design algorithms for 3D joint triangulation and verification through reprojection checks.
- Develop real-time biomechanical analysis tools for posture, movement, and form correction.
- Add temporal filtering and keep camera streams synchronized.
- Tune inference performance to reach latency below 100 ms.
- Support model deployment through ONNX and TensorFlow Lite.
- Set up evaluation datasets, performance measures, and automated testing systems.
Skills and experience needed
- Strong knowledge of computer vision and image processing.
- Hands-on experience with OpenCV.
- Practical understanding of camera calibration, stereo vision, and epipolar geometry.
- Working knowledge of multi-view geometry and 3D reconstruction.
- Experience with feature detection methods such as SIFT, ORB, and AKAZE.
- Ability to apply homography and geometric transformation techniques.
- Experience in human pose estimation and 2D/3D skeleton tracking.
- Comfort with video processing, live streaming, and multi-camera setups.
- Programming experience in Python, NumPy, SciPy, OpenCV, scikit-image, C++, and Git.
- Familiarity with MediaPipe, YOLO, HRNet, ONNX, TensorFlow Lite, Edge AI, and on-device inference.
- Experience with multi-camera or stereo camera systems is preferred.
- Prior deployment of computer vision products in production environments is preferred.
- Exposure to edge AI, mobile deployment, AR/VR, robotics, sports analytics, or fitness technology is a plus.
- Reading familiarity with CVPR, ICCV, or ECCV research papers is preferred.
Work context
This position is based in Bengaluru, India and is with a deep-tech startup focused on real-time perception and decision systems built using computer vision, machine learning, and 3D technologies.
Additional information
No additional compensation details, openings count, start date, duration, or applicant eligibility were provided in the source.