Overview
What this challenge is about.
Train NeRFs on 3 apartment datasets, render novel views, benchmark GPU cost, and run a user study. Deliver a go/no-go memo. Get a verifiable certificate.
The scenario
The proptech (Series A, around 60 staff) currently photographs around 800 high-end listings per year in Berlin, Munich, and Hamburg; each 360 tour costs about EUR 350 to produce and a NeRF-based product would need to land near EUR 200 to be worth the build.
The Brief
What you'll do, and what you'll demonstrate.
Evaluate whether Neural Radiance Fields are production-viable for real-estate virtual tours on cost, quality, and training time.
Earning criteria — what you'll demonstrate
- Understand and apply neural scene representations (NeRF family)
- Design a fair cost + quality benchmark between two production options
- Run a lightweight user study and report results honestly
- Translate research outputs into a business recommendation
Program Fit
Where this fits in your program.
Sharpens the same skills your degree expects you to demonstrate.
Aligned coursework coming soon.
Skills
Skills you'll demonstrate.
Each one shows up on your verified credential.
- Neural Scene Representation
Apply neural scene representation to solve real industry problems and demonstrate production-level capability.
- Nerf
Apply nerf to solve real industry problems and demonstrate production-level capability.
- Pytorch
Apply pytorch to solve real industry problems and demonstrate production-level capability.
- Experiment Design
Apply experiment design to solve real industry problems and demonstrate production-level capability.
- Evaluation
Apply evaluation to solve real industry problems and demonstrate production-level capability.
- Gpu Benchmarking
Apply gpu benchmarking to solve real industry problems and demonstrate production-level capability.
Careers
Career paths this challenge builds toward
Completing this challenge demonstrates skills that transfer directly to these roles: