Super cool work Masha, congrats!

Congrats, Kosta! That sounds incredible. Wishing you an amazing year ahead full of great people, new ideas, and exciting experiences.

Have you already taken off or still around for a bit?

Super insightful!

[9/9] 4Real-Video-V2: Fused View-Time Attention and Feedforward Reconstruction for 4D Scene Generation

🌐 Project page: snap-research.github.io/4Real-Video-V2
📜 Abstract: arxiv.org/abs/2506.18839

[8/9] Authors: Chaoyang Wang*, Ashkan Mirzaei*, Vidit Goel, Willi Menapace, Aliaksandr Siarohin, Avalon Vinella, Michael Vasilkovsky, Ivan Skorokhodov, Vladislav Shakhrai, Sergey Korolev, Sergey Tulyakov, Peter Wonka.
*equal contribution

[7/9] 🧠 We use a camera token replacement trick for temporal consistency of the camera poses, temporal attention layers to share info over time, and a "Gaussian head" to predict shape, scale, opacity, and color offsets.

[6/9] 🔁 How it works – Stage 2 (Reconstruction):
Our feedforward model takes RGB frames and predicts camera poses and dynamic 3D Gaussians. No optimization loops. No ground-truth poses. Just fast, clean reconstruction.

[5/9] ⚡ The architecture runs on a DiT backbone. Thanks to sparse attention and temporal compression, we keep things efficient. Only self-attention layers are fine-tuned and everything else is frozen.

[4/9]🧠How it works – Stage 1 (Generation):
We fuse spatial/temporal attentions into a transformer layer. This view-time attention lets our diffusion model reason across viewpoints and frames jointly, without extra parameters. Parameter-efficiency also leads to more stability.

[3/9] High-quality 4D training data is scarce, and large video models are expensive to fine-tune. So we focus on parameter efficiency. Our fused attention design reuses pretrained weights with minimal changes. It trains fast, generalizes well, and scales to full 4D scenes.

[2/9] We generate synchronized multi-view video grids, then lift them into 4D geometry using a fast feedforward network. The result is a set of Gaussian particles, ready for rendering, exploration, and editing.

📹 EventSplat: 3D Gaussian Splatting from Moving Event Cameras for Real-time Rendering
Toshiya Yura, @ashmrz.bsky.social, Igor Gilitschenski 5/🧵
arxiv.org/abs/2412.07293

Huge congrats Kosta🎉