HDMI (HumanoiD iMitation for Interaction) is a framework enabling humanoid robots to learn…
The Humanoid Hub
在 X 上查看@TheHumanoidHub12万 位关注者
17,443 次观看
HDMI (HumanoiD iMitation for Interaction) is a framework enabling humanoid robots to learn whole-body object interaction skills from monocular RGB human videos.
It extracts and retargets human poses and object trajectories using GVHMR and LocoMujoco, building reference datasets with contact annotations. The data is used to train an RL policy via robot-object co-tracking.
HDMI achieved 67 consecutive door traversals.
10 条评论
评论来自 X (Twitter) 上的原帖

@TheHumanoidHub
More details in this thread
8 个赞

@East_Clipper
This drunk robot again?
3 个赞

@Raph_Pare
I am not drunk.
2 个赞

@ScottDValenti
Awesome... he'll be in every home very soon!
2 个赞

@OriginalTrades
@SolportTom 67

@cryptodaaddy
that's a game changer for robotics and AI integration. using human videos as a learning tool is a smart move

@Solo__Tech
This shows the power of combining human pose retargeting with RL. Robots are starting to internalize not just motion, but intent and interaction.

@san18427
That's me at Saturday 1PM

@AnselDias
Bro you gotta be trolling... HDMI???

@surajgajula
Six Seven