🔥Telegram İfşa

Excited to announce the AI for Math Workshop at #ICML2024 ICML Conference! Join us for groundbreaking discussions on the intersection of AI and mathematics. 🤖🧮 📅 Workshop details: sites.google.com/view/ai4mathwo… 📜 Submit your pioneering work: sites.google.com/view/ai4mathwo… 🏆 Take on our

In Memoriam: Xiaoou Tang | International Journal of Computer Vision #ijcv link.springer.com/article/10.100…

MVSGaussian is super cool. MVSGaussian has several features: 1) high-quality, 2) generalizable and fast training, and 3) real-time rendering. Code is released. Welcome to use our code.

🚀Web-Scale Human Video Dataset🚀 #WildAvatar is a web-scale in-the-wild human avatar creation dataset with 10,000+ diverse human subjects and scenes. Each video is equipped with 🕺SMPL💃 and 🧥mask👚 annotation - Project: wildavatar.github.io - Code: github.com/wildavatar/Wil…

Web-scale human video dataset along with an in-the-wild human video annotation pipeline.

Thanks AK for sharing our latest work, #3DTopiaXL, a native 3D diffusion model for PBR asset generation! 🥳Try out our Hugging Face demo: huggingface.co/spaces/FrozenB… ✨Code release: github.com/3DTopia/3DTopi… Technical report is on its way!

[CV] From Seconds to Hours: Reviewing MultiModal Large Language Models on Comprehensive Long Video Understanding arxiv.org/abs/2409.18938

From Seconds to Hours: Reviewing MultiModal Large Language Models on Comprehensive Long Video Understanding. arxiv.org/abs/2409.18938

Very interesting project, welcome to communicate if you are interested 😀😀😀

Distilled-3DGS:Distilled 3D Gaussian Splatting 1) We propose a novel distillation-based 3DGS framework, termed Distilled-3DGS. It is the first method to leverage multi-teacher knowledge priors to optimize 3DGS and boost rendering quality and storage efficiency. 2) We propose a

Distilled-3DGS:Distilled 3D Gaussian Splatting Lintao Xiang, Xinkai Chen, Jianhuang Lai, Guangcong Wang tl;dr: vanilla 3DGS+noise-augmented 3DGS+dropout-regularized 3DGS->multi teachers->voxel histogram-based structural loss->student arxiv.org/abs/2508.14037

Distilled-3DGS Distilled 3D Gaussian Splatting

12. Distilled-3DGS:Distilled 3D Gaussian Splatting 🔑 Keywords: 3D Gaussian Splatting, Knowledge Distillation, Novel View Synthesis, Structural Similarity Loss 💡 Category: Computer Vision 🌟 Research Objective: - To enhance rendering quality and storage efficiency in 3D

12. Distilled-3DGS:Distilled 3D Gaussian Splatting      蒸馏-3DGS: 蒸馏三维高斯喷射 🔑 关键词: 3D 高斯点, 知识蒸馏, 新视图合成, 结构相似损失 💡 类别: 计算机视觉 🌟 研究目标: - 通过利用知识蒸馏框架提升3D高斯点的渲染质量和存储效率。 🛠️ 研究方法: -

3Dモデル生成時に欠けている部分を動画生成モデルで補う研究「ObjFiller-3D」が登場。やっぱ動画生成モデルは3Dと相性良いよね！ この手の研究はベースモデルの性能向上で不要になる印象があるけど、今すぐ3D生成の品質向上をしたい人には良いアプローチかも？ objfiller3d.github.io

ObjFiller-3D: Consistent Multi-view 3D Inpainting via Video Diffusion Models huggingface.co/papers/2508.18…