Sayan Deb Sarkar

📰 Paper: arxiv.org/abs/2502.15011
▶️ Project Page: sayands.github.io/crossover/
💻 Codebase: github.com/GradientSpaces…

Work w/ Ondrej Miksik, @marcpollefeys.bsky.social, @danielbarath.bsky.social and @ir0armeni.bsky.social ✨

(3/3)

🗓️ Thursday 12 June 3:00 p.m. - 3:45 p.m. CDT
📍 OpenSun3D Workshop Poster Session Arch 211

(2/3)

✨ Excited to head off to Nashville for #CVPR2025

🎤 Catch me at the poster sessions or just come say hi to grab ☕

🗓️ Friday 13 June 4:00 p.m. - 6:00 p.m. CDT
📍 Poster Session #2 — Exhibit Hall D Highlight Poster #346

(1/3)

🥳Excited to share our latest work, WildGS-SLAM: Monocular Gaussian Splatting SLAM in Dynamic Environments, accepted to #CVPR2025 🌐

We present a robust monocular RGB SLAM system that uses uncertainty-aware tracking and mapping to handle dynamic scenes.

🏆 CrossOver is accepted as a 𝗛𝗶𝗴𝗵𝗹𝗶𝗴𝗵𝘁 at #CVPR2025! ✨
💻 Fully open-sourced code with all pre-trained checkpoints: github.com/GradientSpac...

📡 Stay tuned for a deep-dive thread and what else we are cooking 🍳

Looking forward to it!

But, the multimodal problem is same as in image generative tasks — as in, what is the perfect 3D scan given a text input?

In this case, what would be a definitive ground truth?

Thanks for sharing our work! Yes, I think that’d be a pretty neat downstream application but maybe it is more multimodal generation rather than reconstruction.

CrossOver: 3D Scene Cross-Modal Alignment Multi-modal 3D object understanding has gained significant attention, yet current approaches often assume complete data availability and rigid alignment across all modalities. We present CrossOver, a ...

🔗 arXiv: arxiv.org/abs/2502.15011
📂 Project page: sayands.github.io/crossover/

Joint work with Ondrej Miksik, @marcpollefeys.bsky.social, @danielbarath.bsky.social and @ir0armeni.bsky.social 🤝💡

🎉 Excited to share our latest work, CrossOver: 3D Scene Cross-Modal Alignment, accepted to #CVPR2025 🌐✨

We learn a unified, modality-agnostic embedding space, enabling seamless scene-level alignment across multiple modalities — no semantic annotations needed!🚀

CrossOver: 3D Scene Cross-Modal Alignment Multi-modal 3D object understanding has gained significant attention, yet current approaches often assume complete data availability and rigid alignment across all modalities. We present CrossOver, a ...

🔗 arXiv: arxiv.org/abs/2502.15011
📂 Project page: sayands.github.io/crossover/

Joint work with Ondrej Miksik, @marcpollefeys.bsky.social @danielbarath.bsky.social and @ir0armeni.bsky.social 🤝💡

🚀🚀PaliGemma 2 is our updated and improved PaliGemma release using the Gemma 2 models and providing new pre-trained checkpoints for the full cross product of {224px,448px,896px} resolutions and {3B,10B,28B} model sizes.

1/7

Would love to be added! I’m a PhD student working on 3D scene understanding and spatial AI.

Would love to be added! I’m a PhD student working on 3D scene understanding and spatial AI.

Would love to be added! I’m a PhD student working on 3D scene understanding and spatial AI.

Could you add me? I’m a PhD student working on 3D scene understanding.