Remote Computer Vision Engineer — Human Pose Estimation

Xsens Technologies B.V.; Enschede; Hybrid; EUR 70,000 - 95,000; Full time

Benefits offered by this job

28 paid vacation days

Tax perks for sports memberships and a

bicycle purchase;

Flexible working hours

Paid volunteer time

On-demand financial planner access

Company social events and activities

Job summary

Xsens Technologies B.V. is seeking a hands-on Computer Vision Engineer to lead camera-based perception for motion capture. You will build and ship DL pipelines to turn camera images into robust human pose estimates, while collaborating with sensor fusion experts. The role is open to hybrid or remote working.

You will advance 2D/3D pose estimation, calibrate diverse camera setups, and evaluate outputs with real-time considerations in a world-class high-tech environment.

Qualifications

• MSc/PhD in CS/EE/Robotics with years of applied research experience.
• Hands-on experience training/refining deep-learning models for 2D/3D pose estimation.
• Solid camera and lens modelling knowledge, including calibration and multi-view geometry.
• Experience with OpenCV, SAPIENS, SAM3D and pose representations.
• Proficiency in Python; working knowledge of C++ for integration.

Responsibilities

• Select, adapt and fine-tune DL models for 2D/3D human pose estimation in real-world conditions.
• Develop camera models and calibration pipelines for various camera setups.
• Evaluate motion capture output and develop compensation for vision failure modes.
• Build training/evaluation pipelines from camera and optical motion capture data.
• Stay ahead of CV research trends to assess feasibility for Xsens products.

Skills

Deep learning

Pose estimation

Python

C++

OpenCV

SAPIENS

SAM3D

Multi-view geometry

Camera models

Calibration

Education

MSc/PhD in Computer Science, Electrical Engineering, Robotics

Tools

PyTorch

ONNX

TensorRT

Job description

Xsens Technologies B.V. is seeking a hands-on Computer Vision Engineer to lead camera-based perception for motion capture. You will build and ship DL pipelines to turn camera images into robust human pose estimates, while collaborating with sensor fusion experts. The role is open to hybrid or remote working.

You will advance 2D/3D pose estimation, calibrate diverse camera setups, and evaluate outputs with real-time considerations in a world-class high-tech environment.