XR Object Localization with World Models (Master Thesis)

Ericsson; Stockholms kommun; On-site; SEK 39,000 - 61,000; Full time

Job summary

Ericsson is seeking a Master’s student to investigate world-model-based depth and motion estimation for monocular XR video. The project involves building a Quest-ready capture pipeline with server-side inference and a lightweight distance/projection probe, evaluated against depth sensors.

You will develop a prototype, document datasets, and produce results with recommendations for camera-only XR perception and network-assisted spatial intelligence.

Qualifications

• Master’s degree in XR or spatial understanding, currently enrolled.
• Fluent written and spoken English.
• Experience with Unity and MR/VR programming.
• Experience writing apps for the Meta Quest family is a plus.
• Committed to Ericsson’s values: Professionalism, Perseverance, Respect, Integrity.

Responsibilities

• Build headset RGB-video capture pipeline with server-side inference.
• Train a lightweight probe to estimate object distance and motion while keeping backbone frozen.
• Evaluate the system in controlled scenes with moving objects using ground-truth baselines.
• Assess accuracy, computational requirements, latency, and uplink needs.
• Develop a working prototype for XR perception research.
• Create a documented dataset and measurement setup for reproducibility.
• Present quantitative accuracy and latency results in the thesis report.
• Provide recommendations for camera-only XR perception and network-assisted spatial intelligence.

Skills

Analytical

English fluency

XR/spatial understanding

Unity

MR/VR programming

Meta Quest apps

Education

Master’s degree in XR/spatial understanding

Tools

Unity

Job description

Ericsson is seeking a Master’s student to investigate world-model-based depth and motion estimation for monocular XR video. The project involves building a Quest-ready capture pipeline with server-side inference and a lightweight distance/projection probe, evaluated against depth sensors.

You will develop a prototype, document datasets, and produce results with recommendations for camera-only XR perception and network-assisted spatial intelligence.