Librarian Bot. I found the following papers similar to this paper.

The following papers were recommended by the Semantic Scholar API

PoCo: Policy Composition from and for Heterogeneous Robot Learning (2024)
CyberDemo: Augmenting Simulated Human Demonstration for Real-World Dexterous Manipulation (2024)
General Flow as Foundation Affordance for Scalable Robot Learning (2024)
Point Cloud Matters: Rethinking the Impact of Different Observation Spaces on Robot Learning (2024)
Towards Diverse Behaviors: A Benchmark for Imitation Learning with Human Demonstrations (2024)

Please give a thumbs up to this comment if you found it helpful!

If you want recommendations for any Paper on Hugging Face checkout this Space

You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: \n\n@librarian-bot\n\t recommend

\n","updatedAt":"2024-03-08T01:21:57.283Z","author":{"_id":"63d3e0e8ff1384ce6c5dd17d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg","fullname":"Librarian Bot (Bot)","name":"librarian-bot","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":264}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7626360654830933},"editors":["librarian-bot"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2403.03954","authors":[{"_id":"65e9558e6434c9e6f4a30604","user":{"_id":"63509bc859bfa9a85d4220aa","avatarUrl":"/avatars/ca2cc9b87f5ca5cd51606b2f9edf89d0.svg","isPro":false,"fullname":"Yanjie Ze","user":"yjze","type":"user"},"name":"Yanjie Ze","status":"claimed_verified","statusLastChangedAt":"2024-03-07T09:13:40.115Z","hidden":false},{"_id":"65e9558e6434c9e6f4a30605","user":{"_id":"65e95effe19214e9d13d7a2a","avatarUrl":"/avatars/35ae40579ac53d7b598957be9c8b4710.svg","isPro":false,"fullname":"Gu Zhang","user":"GuZhang","type":"user"},"name":"Gu Zhang","status":"claimed_verified","statusLastChangedAt":"2024-03-07T09:13:37.901Z","hidden":false},{"_id":"65e9558e6434c9e6f4a30606","user":{"_id":"63b6def76fca9d2a1902fa14","avatarUrl":"/avatars/c7f2487450ea954e2bca4fc5a6db8eb3.svg","isPro":false,"fullname":"张康宁","user":"zhuiguang-ning","type":"user"},"name":"Kangning Zhang","status":"claimed_verified","statusLastChangedAt":"2024-03-07T16:02:46.137Z","hidden":false},{"_id":"65e9558e6434c9e6f4a30607","user":{"_id":"63ec3830ca08a72ba9d04b81","avatarUrl":"/avatars/12245876937e12651dc794ca99f9109d.svg","isPro":false,"fullname":"chenyuan hu","user":"hcy1223","type":"user"},"name":"Chenyuan Hu","status":"admin_assigned","statusLastChangedAt":"2024-03-07T09:49:28.551Z","hidden":false},{"_id":"65e9558e6434c9e6f4a30608","user":{"_id":"656492f7055bb4df5e8a3a49","avatarUrl":"/avatars/b3b7e8e10853541a9bfa1ce7ce2217fd.svg","isPro":false,"fullname":"Muhan Wang","user":"Sparklist","type":"user"},"name":"Muhan Wang","status":"admin_assigned","statusLastChangedAt":"2024-03-07T09:49:34.997Z","hidden":false},{"_id":"65e9558e6434c9e6f4a30609","user":{"_id":"649c130f92525b3ac9fbcffe","avatarUrl":"/avatars/b60e4d6803f7e4ef42200ddcb9e5bac1.svg","isPro":false,"fullname":"Huazhe Xu","user":"Huazhe-Xu","type":"user"},"name":"Huazhe Xu","status":"admin_assigned","statusLastChangedAt":"2024-03-07T09:49:41.282Z","hidden":false}],"publishedAt":"2024-03-06T18:58:49.000Z","submittedOnDailyAt":"2024-03-07T03:20:09.693Z","title":"3D Diffusion Policy","submittedOnDailyBy":{"_id":"60f1abe7544c2adfd699860c","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674929746905-60f1abe7544c2adfd699860c.jpeg","isPro":false,"fullname":"AK","user":"akhaliq","type":"user"},"summary":"Imitation learning provides an efficient way to teach robots dexterous\nskills; however, learning complex skills robustly and generalizablely usually\nconsumes large amounts of human demonstrations. To tackle this challenging\nproblem, we present 3D Diffusion Policy (DP3), a novel visual imitation\nlearning approach that incorporates the power of 3D visual representations into\ndiffusion policies, a class of conditional action generative models. The core\ndesign of DP3 is the utilization of a compact 3D visual representation,\nextracted from sparse point clouds with an efficient point encoder. In our\nexperiments involving 72 simulation tasks, DP3 successfully handles most tasks\nwith just 10 demonstrations and surpasses baselines with a 55.3% relative\nimprovement. In 4 real robot tasks, DP3 demonstrates precise control with a\nhigh success rate of 85%, given only 40 demonstrations of each task, and shows\nexcellent generalization abilities in diverse aspects, including space,\nviewpoint, appearance, and instance. Interestingly, in real robot experiments,\nDP3 rarely violates safety requirements, in contrast to baseline methods which\nfrequently do, necessitating human intervention. Our extensive evaluation\nhighlights the critical importance of 3D representations in real-world robot\nlearning. Videos, code, and data are available on\nhttps://3d-diffusion-policy.github.io .","upvotes":14,"discussionId":"65e955916434c9e6f4a3071c","ai_summary":"3D Diffusion Policy (DP3) utilizes 3D visual representations for efficient imitation learning, achieving robust and generalized skills with fewer demonstrations compared to existing methods.","ai_keywords":["3D Diffusion Policy","DP3","3D visual representations","sparse point clouds","point encoder","diffusion policies","conditional action generative models"]},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"63509bc859bfa9a85d4220aa","avatarUrl":"/avatars/ca2cc9b87f5ca5cd51606b2f9edf89d0.svg","isPro":false,"fullname":"Yanjie Ze","user":"yjze","type":"user"},{"_id":"65e95effe19214e9d13d7a2a","avatarUrl":"/avatars/35ae40579ac53d7b598957be9c8b4710.svg","isPro":false,"fullname":"Gu Zhang","user":"GuZhang","type":"user"},{"_id":"6362ddb7d3be91534c30bfd6","avatarUrl":"/avatars/dac76ebd3b8a08099497ec0b0524bc7c.svg","isPro":false,"fullname":"Art Atk","user":"ArtAtk","type":"user"},{"_id":"635964636a61954080850e1d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/635964636a61954080850e1d/0bfExuDTrHTtm8c-40cDM.png","isPro":false,"fullname":"William Lamkin","user":"phanes","type":"user"},{"_id":"62583d8189aaec1123517d9c","avatarUrl":"/avatars/30b6cf0c9af5e3e3804e6b8c6f916a8a.svg","isPro":false,"fullname":"Maxim Bobrin","user":"pupsicated","type":"user"},{"_id":"63b6def76fca9d2a1902fa14","avatarUrl":"/avatars/c7f2487450ea954e2bca4fc5a6db8eb3.svg","isPro":false,"fullname":"张康宁","user":"zhuiguang-ning","type":"user"},{"_id":"63d4c8ce13ae45b780792f32","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1675778487155-63d4c8ce13ae45b780792f32.jpeg","isPro":false,"fullname":"Ohenenoo","user":"PeepDaSlan9","type":"user"},{"_id":"6538119803519fddb4a17e10","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6538119803519fddb4a17e10/ffJMkdx-rM7VvLTCM6ri_.jpeg","isPro":false,"fullname":"samusenps","user":"samusenps","type":"user"},{"_id":"6108956e7602f8e9ed8bb5d8","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1672966209331-6108956e7602f8e9ed8bb5d8.png","isPro":false,"fullname":"adakoda","user":"adakoda","type":"user"},{"_id":"6303e34ed14428368d192f45","avatarUrl":"/avatars/f02b5b457f818e87b3ab8c1341b12062.svg","isPro":true,"fullname":"Clayface","user":"ClayFace","type":"user"},{"_id":"63652bec12188d67e656644c","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63652bec12188d67e656644c/-czLQBRI6QaKSGzjCWy2j.png","isPro":false,"fullname":"ramune","user":"ramu0e","type":"user"},{"_id":"6788957ceb36144551bc9bc7","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6788957ceb36144551bc9bc7/PsNCHhimfKSNH3IgthXoF.png","isPro":false,"fullname":"LiangJunCHEN","user":"KOKOMIA","type":"user"}],"acceptLanguages":["*"],"dailyPaperRank":0}">

Papers

arxiv:2403.03954

3D Diffusion Policy

Published on Mar 6, 2024

· Submitted by

AK on Mar 7, 2024

Upvote

Authors:

Yanjie Ze ,

Gu Zhang ,

Kangning Zhang ,

Chenyuan Hu ,

Muhan Wang ,

Huazhe Xu

Abstract

3D Diffusion Policy (DP3) utilizes 3D visual representations for efficient imitation learning, achieving robust and generalized skills with fewer demonstrations compared to existing methods.

AI-generated summary

Imitation learning provides an efficient way to teach robots dexterous skills; however, learning complex skills robustly and generalizablely usually consumes large amounts of human demonstrations. To tackle this challenging problem, we present 3D Diffusion Policy (DP3), a novel visual imitation learning approach that incorporates the power of 3D visual representations into diffusion policies, a class of conditional action generative models. The core design of DP3 is the utilization of a compact 3D visual representation, extracted from sparse point clouds with an efficient point encoder. In our experiments involving 72 simulation tasks, DP3 successfully handles most tasks with just 10 demonstrations and surpasses baselines with a 55.3% relative improvement. In 4 real robot tasks, DP3 demonstrates precise control with a high success rate of 85%, given only 40 demonstrations of each task, and shows excellent generalization abilities in diverse aspects, including space, viewpoint, appearance, and instance. Interestingly, in real robot experiments, DP3 rarely violates safety requirements, in contrast to baseline methods which frequently do, necessitating human intervention. Our extensive evaluation highlights the critical importance of 3D representations in real-world robot learning. Videos, code, and data are available on https://3d-diffusion-policy.github.io .