Ever wondered how a scene sounds👂 when you interact👋 with it?. Introducing our #CVPR2025 work "Hearing Hands: Generating Sounds from Physical Interactions in 3D Scenes" -- we make 3D scene reconstructions audibly interactive! .

RT @pliang279: Despite much progress in AI, the ability for AI to 'smell' like humans remains elusive. Smell AIs 🤖👃can be used for allergen….

RT @jin_linyi: Hello! If you are interested in dynamic 3D or 4D, don't miss the oral session 3A at 9 am on Saturday:. @zhengqi_li .will be….

RT @ayshrv: Excited to share our CVPR 2025 paper on cross-modal space-time correspondence!. We present a method to match pixels across diff….

RT @jespark0: Can AI image detectors keep up with new fakes?. Mostly, no. Existing detectors are trained using a handful of models. But the….

Wonderful collaboration with Wonseok Oh, Yuqing Luo, @antoniloq, @andrewhowens!!!.

Combining with our previous #CVPR2024 work TaRF (, we create an immersive 3D scene reconstruction that allows users to interact with it using sight👀, touch👆 and sound👂.

Next, we train a rectified flow model to generate the interaction sound conditioned on a 3D hand trajectory and video frames rendered from a 3D reconstruction of a scene. Generally, the predictions are accurate in both motion synchronization and material properties.

To capture the interactions, the human collector interacts with the scene by performing various actions with their hands. We lift the annotator's hands to the scene reconstruction, and render a video of the interaction by projecting 3D hands on the scene.

Traditional 3D reconstruction captures realistic visuals, but what about the sounds of interactions? Our work bridges this gap by predicting realistic audio from hand-object interactions in 3D scenes.

RT @dangengdg: Hello! If you like pretty images and videos and want a rec for CVPR oral session, you should def go to Image/Video Gen, Frid….

RT @_crockwell: Ever wish YouTube had 3D labels?. 🚀Introducing🎥DynPose-100K🎥, an Internet-scale collection of diverse videos annotated with….

RT @ju_yuanchen: 🧩#CVPR2025🌷Introducing Two By Two✌️: The First Large-Scale Daily Pairwise Assembly Dataset with SE(3)-Equivariant Pose Est….

Thanks to @OpenAI, got a chance to grow up again in Ghibli anime🤗

RT @SarahJabbour_: I’m on the PhD internship market for Spr/Summer 2025! I have experience in multimodal AI (EHR, X-ray, text), explainabil….

RT @ju_yuanchen: 🍌We present DenseMatcher！.🤖️DenseMatcher enables robots to acquire generalizable skills across diverse object categories b….

RT @dangengdg: What happens when you train a video generation model to be conditioned on motion?. Turns out you can perform "motion prompti….

RT @junyi42: Excited to share MonST3R! -- a simple way to estimate geometry from unposed video of dynamic scene. We achieve competitive res….

RT @Zichen2501: Differentiable rendering made SIMPLE❗️. Differentiating physically based renderers is hard: Dirac-delta discontinuities ari….

RT @ayshrv: We present Global Matching Random Walks, a simple self-supervised approach to the Tracking Any Point (TAP) problem, accepted to….