Associate Research Professor at @SCSatCMU, @CMU_Robotics. Computer Vision, Machine Learning and Accessibility.

Pittsburgh, PA
Joined June 2009
Kris Kitani retweeted
Generated 3D parts can look great—but can they form a physical object? We dropped 100 assemblies under gravity. 0 stayed stable. With SNAP3D, 95% do. We make part-aware 3D generation physically grounded with geometry repair + physics-optimized connectors.
2
6
2
17
5,487
Kris Kitani retweeted
❓ What should 3D part decomposition look like? 📣 Introducing Point2Part, a unified model that treats 3D part decomposition as a partition of the whole, rather than a collection of independently predicted parts. Point2Part lets users specify the parts they want with 3D points and produces a complete, non-overlapping decomposition of the whole shape. One model supports image→parts, mesh→parts, and mesh segmentation, achieving SOTA part quality with >10× lower inter-part penetration. 📄 arXiv: arxiv.org/abs/2609.38180 🌐 Project: henrytsui000.github.io/Point… Joint work with @LucyTuan1, @XiaoxuanMa_, @nicougrinovic, Takaaki Shiratori, and @kkitani.
1
15
1
32
12,415
20+ new papers from CMU at CVPR 2026!!! vision.cs.cmu.edu/papers.htm…
5
44
4,041
A snapshot into how SAM 3D Body is being used at CMU for assisting physical therapy
SAM 3D is helping advance the future of rehabilitation. See how researchers at @CarnegieMellon are using SAM 3D to capture and analyze human movement in clinical settings, opening the doors to personalized, data-driven insights in the recovery process. 🔗 Learn more about SAM 3D: go.meta.me/305985
5
1
55
7,852
Sam3 + Body is freaking amazing. I'm in the process of building an open-source @rerundotio and @Gradio demo that is similar to what Meta provided. Got the basic functionality up and running, now I need to hook it up to a Gradio interface. It's a really good model
Egocentric slam collection on my own hardware achieved with @rerundotio! Previously, I'd all my own ego collections used a static camera. This is obviously not a good long-term solution and only really made sense for initial development. Now I can easily collect samples that have super-accurate slam from a quest device, and the giant field of view from a stereo + RGB camera rig. As you can see in the video, I'm still working through some bugs, but I wanted to share some progress anyway. Hoping to have all kinks worked out by Friday. The main thing I need to fix is a weird offset happening between each of the rigs. I need to stop representing cameras as individuals, but rather as rigs that have fixed relative positions to ensure rigid motion.
12
69
4
711
122,872
Use 3DB for better camera control! Uni3C + SAM 3D: The Ultimate Workflow for AI Camera Control|3D Motion wi... youtu.be/pXiE5yjX4K0?si=VYBu… via @YouTube
8
19
6,229
Kris Kitani retweeted
RustDesk神すぎる… 添付は流行りの3Dモデリングできるsam-3d-bodyで星街さんをWindowsで3D化しMacのBlenderで表示したもの Mac Loveの私にはWindows環境は辛すぎるのでMac で Windows実機をP2Pでそのまま操作できるのがありがたすぎます キーバインドでctrlを⌘キーにもできるので画面スワイプしながらMacの1画面として操作できる神ツール ずっとMagic Trackpadで操作したかったのです... ちなみにsam-3d-bodyはwindowsだとEGL、detectron2の問題で実行できないので(多分) detectをカットするので本来のパワーを発揮できないのですが実際のobjとして生成自体は可能でした 素の人体模型になってしまうので取り扱いがちょっとむつかしい
2
1
18
1,855
SAM 3D × NanoBanana2 × Kling = 位置コントロール 🐰❤️‍🔥 ワークフロー: 1️⃣ SAM 3D Objects でモチーフを3D化 2️⃣ 好きな順番・配置に並び替える 3️⃣ Nano Banana 2 で新しい画像を生成 4️⃣ 仕上げに Kling で前後フレームの動画化 結果は—— ✨位置の再現精度ほぼ90%!✨ 🐰❤️‍🔥心臓バクバクしました。これは本当にすごい!!新しい創作の自由そのものです。 2025.11.21 @AIatMeta @GeminiApp @Kling_ai #MetaAI #klingai #SAM3D #NanoBanana2 #kanaworksai
2
13
1
72
3,260
Kris Kitani, faculty in @CMU_Robotics, led the development of SAM3D Body, which could transform physical therapy with AI pose estimation. SAM 3D Body delivers precise, AI-driven static position analysis to personalize rehabilitation and enhance patient outcomes.
Introducing SAM 3D, the newest addition to the SAM collection, bringing common sense 3D understanding of everyday images. SAM 3D includes two models: 🛋️ SAM 3D Objects for object and scene reconstruction 🧑‍🤝‍🧑 SAM 3D Body for human pose and shape estimation Both models achieve state-of-the-art performance transforming static 2D images into vivid, accurate reconstructions. 🔗 Learn more: go.meta.me/305985
2
3
17
2,777
Transforming physical therapy with AI pose estimation. SAM 3D Body delivers precise, AI-driven static position analysis to personalize rehabilitation and enhance patient outcomes.
1
1
1
7
790
SAM 3D × Nano Banana × Kling × Domo のコンボが楽しすぎる…!🐰🤓 指定した人物を好きなアングル&ポーズで3D化 → 動画生成 → そのまま画風チェンジまで 一気通貫でいけるのヤバい🤣 本当に「自分専用の3D&動画スタジオ」って感じ。 2025.11.20 @AIatMeta @Kling_ai @DomoAI_ #SAM3D #NanoBanana #DomoAI #AI動画 #kanaworksai #MetaAI #klingai #KlingLab
5
75
3
507
35,179
Meta AI のSAM 3D |テスト② SAM 3D Body これはもう狂ってるレベル😱🥳🍻 ・単一画像→全身の形&ポーズを3D再構築 ・一部が隠れていてもちゃんと補完 ・人物の変形&インタラクション向き 2025.11.20 @AIatMeta  #SAM3D #MetaAI #3Dモデリング #Blender #ゲーム開発 #b3d #kanaworksai
Meta AI の最新モデル SAM 3D の SAM 3D Objects を試してみたら、本当に衝撃的でした。 たった1枚の画像から、物体の 3D形状・質感・空間レイアウト まで復元してくれるなんて…すごすぎる。 AI はもう「画像を見る」だけじゃなく、「空間を理解し始めている」んだなぁ。 一晩寝て起きたら、世界がアップデートされてた気分🤣😱😍 🐰ワクワク 2025.11.20 #MetaAI #SAM3D
4
35
5
178
23,901
📢 Recover highly accurate full-body pose and shape with perfect 2D alignment! Very excited to finally share SAM 3D Body, work done at Meta! Tech report: ai.meta.com/research/publica… Demo: aidemos.meta.com/segment-any… Blog: ai.meta.com/blog/sam-3d/
1
4
505
Kris Kitani retweeted
We’re sharing model checkpoints, an evaluation benchmark, human body training data, and inference code with the community to support creative applications in fields like robotics, interactive media, science, sports medicine, and beyond. 🔗 SAM 3D Body: github.com/facebookresearch/… 🔗 SAM 3D Objects: github.com/facebookresearch/…
9
13
2
243
39,921
Kris Kitani retweeted
Introducing SAM 3D, the newest addition to the SAM collection, bringing common sense 3D understanding of everyday images. SAM 3D includes two models: 🛋️ SAM 3D Objects for object and scene reconstruction 🧑‍🤝‍🧑 SAM 3D Body for human pose and shape estimation Both models achieve state-of-the-art performance transforming static 2D images into vivid, accurate reconstructions. 🔗 Learn more: go.meta.me/305985
127
1,040
301
6,350
861,841
New model from Meta, SAM 3D Body, powered by people from Smith Hall (@kkitani,@jinkuncao, David Park, Jyun-Ting Song) of course! #goSmithHall Introducing SAM 3D: a New Standard for 3D Object & Human Reconstruction ... youtu.be/B7PZuM55ayc?si=dOBJ… via @YouTube
2
8
1,079
Super excited to share the release of SAM 3D. It's been a year in the making. Two models for lifting object and people to 3D!
Introducing SAM 3D, the newest addition to the SAM collection, bringing common sense 3D understanding of everyday images. SAM 3D includes two models: 🛋️ SAM 3D Objects for object and scene reconstruction 🧑‍🤝‍🧑 SAM 3D Body for human pose and shape estimation Both models achieve state-of-the-art performance transforming static 2D images into vivid, accurate reconstructions. 🔗 Learn more: go.meta.me/305985
8
11
165
15,709
Kris Kitani retweeted
Meet BFM-Zero: A Promptable Humanoid Behavioral Foundation Model w/ Unsupervised RL👉 lecar-lab.github.io/BFM-Zero… 🧩ONE latent space for ALL tasks ⚡Zero-shot goal reaching, tracking, and reward optimization (any reward at test time), from ONE policy 🤖Natural recovery & transition
7
75
23
271
84,167
Excited to release our new preprint - we introduce Adaptive Patch Transformers (APT), a method to speed up vision transformers by using multiple different patch sizes within the same image!
10
29
12
228
30,313
Kris Kitani retweeted
To help fuel new research in the community around video learning and egocentric perception, together with the Ego4D consortium we’ve published a new research paper covering our work on Ego-Exo4D. Access the full paper ➡️ bit.ly/3NxdlhL
11
41
5
283
55,848