lossfunk resident building open-source 3D human avatar pipeline from video
paraschopra · x · 2026-10-27
Vaibhav, a lossfunk Batch 9 resident, is building a pipeline that extracts high-fidelity human avatars with accurate 3D positioning from ordinary video streams. Milestones: curating a dataset annotated with mesh, skeleton poses and spatial labels, implementing and training the network in tinygrad, then iterating. Both the dataset and model will be open-sourced.
More from Companies & People
- Sequoia, Citi and OpenAI launch No Limit: build a poker bot in 60 minutes, $100K grand prize — polynoamial · 2026-10-28
- The only moat left: building businesses led by people with integrity — dr_alphalyrae · 2026-10-28
- Formic CEO Co-Writes Essay on the Realistic Path to General-Purpose Robots That Don't Suck — AnneliesGamble · 2026-10-28
- Sequoia and OpenAI host $100K poker bot tournament where agents build bots in 60 minutes — polynoamial · 2026-10-28
- PPCC26 demos praised: real orgs like NFL, real code live in VS Code, no fake companies — donasarkar · 2026-10-28
- NYT's Kevin Roose Jokes That 5 of Anthropic's 7 Co-founders 'Got New Faces' in the S-1 — kevinroose · 2026-10-28