AI Paper Atlas

Research map / 3D vision / 3D generation

UGG: Unified Generative Grasping

Jiaxin Lu, Hao Kang, Haoxiang Li, Bo Liu, Yiding Yang, Qixing Huang, Gang Hua

ECCV 2024 · 3D vision › 3D generation · approach: Hand-object interaction generation · 44 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add physics discriminator to select top candidates (ablation)UGG · DexGraspNetsuccess rate: 64.1 -> 72.7Improved
add physics discriminator to select top candidates (ablation)UGG · DexGraspNetjoint entropy H mean (diversity): 8.31 -> 7.17Hurt
add contact anchor and hand encoder (ablation)UGG · DexGraspNet 20-object subsetsuccess rate: 54.3 -> 59.9Improved
add hand encoder alone (no contact anchor) (ablation)UGG · DexGraspNet 20-object subsetsuccess rate and diversity: success 54.3 -> 53.5; H mean 7.92 -> 7.98No clear effect
add contact anchor on top of hand encoder (ablation)UGG · DexGraspNet 20-object subsetQ1: 0.072 -> 0.037Hurt
use contact map instead of contact anchors in the joint generative model (ablation)UGG · joint generationgenerated contact map qualityHurt
select more poses by ranking discriminator scores (ablation)physics discriminator · DexGraspNet 1000 generated posessuccess rateHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy