Robotics

C2Dex Turns Phone Videos Into Dexterous Hand Trajectories

Shar Hendrix 5 min read

Collecting clean demos for multi-finger hands is expensive. Phone video of people manipulating objects is not โ€” if you can turn noisy monocular hand-object footage into something a robot hand can actually run.

C2Dex, submitted to arXiv on Aug 7, 2026 (arXiv:2608.07045) and aimed at IEEE RA-L, is a video-to-dexterous-manipulation pipeline from Nanjing University and collaborators. The core trick is not better per-frame pose estimation. It is stable object-side contacts: aggregate noisy frame-wise contact observations into the canonical object frame, then use those contacts both to clean the human trajectory and to retarget it onto a robot hand.

C2Dex overview from monocular video to dexterous hand trajectory
Pipeline overview: contact-consistent HOI reconstruction, then interaction-preserving retargeting. Source: arXiv:2608.07045.

What they built

C2Dex has two coupled modules:

  • Contact-consistent HOI reconstruction โ€” start from Dyn-HaMR hand estimates, SAM 3D object geometry, and ProxyPose object poses; filter contacts with silhouette overlap and normal consistency; cluster contacts across stable temporal segments with DBSCAN; refine MANO trajectories with contact, SDF penetration, and regularization losses.
  • Contact-interaction-preserving retargeting โ€” transfer stable contacts to the target hand (Inspire in the paper), preserve local hand-object geometry with Laplacian interaction optimization, then refine in Isaac Gym with residual RL (ManipTrans-style).

The project page (k-jie.github.io/C2Dex) shows open-loop replay on a Unitree G1 with an Inspire hand on eight contact-rich tasks: hang cap, pour juice, sweep litter, wipe board, stack cups, and more. Twenty-four human demos feed the pipeline with no per-task policy training and no teleop on the robot.

C2Dex method diagram for HOI reconstruction and retargeting
Method diagram from the project page. Source: C2Dex project site / arXiv:2608.07045.

Numbers that matter

Under identical evaluation criteria on monocular demos:

DatasetC2Dex (relaxed)Strongest baseline
DexYCB (45 seq)57.78% (26/45)17.78% (Do As I Do)
TACO (30 seq)26.67% (8/30)10.00% (Do As I Do)

Strict rotation thresholds still leave C2Dex far ahead (55.56% / 23.33%). On retargeting alone with ground-truth human HOI input, max penetration on DexYCB drops from 22.92 mm (DexPilot) to 3.99 mm. Ablations show removing cross-frame contact consistency collapses DexYCB success back to 17.78%.

Real-world C2Dex replay tasks hang cap sweep wipe drop litter
Real-robot open-loop replay on hang, sweep, wipe, and drop tasks. Source: arXiv:2608.07045.

A Humanโ€™s Take

Iโ€™m so here for the boring object-space contact bank. Most video-to-dexterity stacks fail because contacts jitter every frame and retargeting matches fingertips instead of where the object actually gets touched. If C2Dex-style stable contacts hold up outside author-recorded demos and clean object meshes, phone footage stops being a research novelty and starts looking like a data factory for hands.

Sources