paper-with-me

홈 › Papers

GS-DiT: Advancing Video Generation with Dynamic 3D Gaussian Fields through Efficient Dense 3D Point Tracking

2025-01-01 · CVPR 2025 1 · Weikang Bian, Zhaoyang Huang, Xiaoyu Shi, Yijin Li, Fu-Yun Wang, Hongsheng Li

4D video control is essential in video generation as it enables the use of sophisticated lens techniques, such as multi-camera shooting and dolly zoom, which are currently unsupported by existing methods. Training a video Diffusion Transformer (DiT) directly to control 4D content requires expensive multi-view videos. Inspired by Monocular Dynamic novel View Synthesis (MDVS) that optimizes a 4D representation and renders videos according to different 4D elements, such as camera pose and object motion editing, we bring dynamic 3D Gaussian fields to video generation. Specifically, we propose a novel framework that constructs dynamic 3D Gaussian fields with dense 3D point tracking and renders the Gaussian field for all video frames. Then we finetune a pretrained DiT to generate videos following the guidance of the rendered video, dubbed as GS-DiT. To boost the training of the GS-DiT, we also propose an efficient Dense 3D Point Tracking (D3D-PT) method for the dynamic 3D Gaussian field construction. Our D3D-PT outperforms SpatialTracker, the state-of-the-art sparse 3D point tracking method, in accuracy and accelerates the inference speed by two orders of magnitude. During the inference stage, GS-DiT can generate videos with the same dynamic content while adhering to different camera parameters, addressing a significant limitation of current video generation models. GS-DiT demonstrates strong generalization capabilities and extends the 4D controllability of Gaussian splatting to video generation beyond just camera poses. It supports advanced cinematic effects through the manipulation of the Gaussian field and camera intrinsics, making it a powerful tool for creative video production. Demos are available at https://wkbian.github.io/Projects/GS-DiT/.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Tasks

Novel View SynthesisPoint TrackingVideo Generation

Methods 이 논문이 사용한 방법론

Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Adam 설명 없음
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
SPEED The monocular depth estimation (MDE) is the task of estimating depth from a single frame. This information is an essential knowledge in many computer vision tasks such as scene…
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Diffusion Diffusion models generate samples by gradually removing noise from a signal, and their training objective can be expressed as a reweighted variational lower-bound…
Position-Wise Feed-Forward Layer 설명 없음
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…

Similar Papers 제목 키워드 기반

GS-DiT: Advancing Video Generation with Pseudo 4D Gaussian Fields through Efficient Dense 3D Point Tracking

2025-01-05 · Weikang Bian, Zhaoyang Huang, Xiaoyu Shi, Yijin Li 외

4D video control is essential in video generation as it enables the use of sophisticated lens techniques, such as multi-camera shooting and dolly zoom, which are currently unsupported by existing methods. Training a vide…

Novel View SynthesisPoint TrackingVideo Generation

Learning Physics-Grounded 4D Dynamics with Neural Gaussian Force Fields

2026-01-29 · Shiqian Li, Ruihong Shen, Junfeng Ni, Chang Pan 외 arxiv

Predicting physical dynamics from raw visual data remains a major challenge in AI. While recent video generation models have achieved impressive visual quality, they still cannot consistently generate physically plausibl…

Video GenerationVideo Prediction

GaussianFlow: Splatting Gaussian Dynamics for 4D Content Creation

2024-03-19 · Quankai Gao, Qiangeng Xu, Zhe Cao, Ben Mildenhall 외

Creating 4D fields of Gaussian Splatting from images or videos is a challenging task due to its under-constrained nature. While the optimization can draw photometric reference from the input videos or be regulated by gen…

Novel View SynthesisOptical Flow Estimation

3D Gaussian Representations with Motion Trajectory Field for Dynamic Scene Reconstruction

2025-08-10 · Xuesong Li, Lars Petersson, Vivien Rolland arxiv

This paper addresses the challenge of novel-view synthesis and motion reconstruction of dynamic scenes from monocular video, which is critical for many robotic applications. Although Neural Radiance Fields (NeRF) and 3D …

Dynamic Camera Poses and Where to Find Them

2025-01-01 · CVPR 2025 1 · Chris Rockwell, Joseph Tung, Tsung-Yi Lin, Ming-Yu Liu 외

Annotating camera poses on dynamic Internet videos at scale is critical for advancing fields like realistic video generation and simulation. However, collecting such a dataset is difficult, as most Internet videos ar…

Point TrackingPose EstimationVideo Generation