VD3D: Taming Large Video Diffusion Transformers for 3D Camera Control

• URL: http://arxiv.org/abs/2407.12781v2
• Date: Sat, 20 Jul 2024 19:43:10 GMT
• Title: VD3D: Taming Large Video Diffusion Transformers for 3D Camera Control
• Authors: Sherwin Bahmani, Ivan Skorokhodov, Aliaksandr Siarohin, Willi Menapace, Guocheng Qian, Michael Vasilkovsky, Hsin-Ying Lee, Chaoyang Wang, Jiaxu Zou, Andrea Tagliasacchi, David B. Lindell, Sergey Tulyakov,
• Abstract summary: We tame transformers video for 3D camera control using a ControlNet-like conditioning mechanism based on Plucker coordinates. Our work is the first to enable camera control for transformer-based video diffusion models.
• Score: 74.5434726968562
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Modern text-to-video synthesis models demonstrate coherent, photorealistic generation of complex videos from a text description. However, most existing models lack fine-grained control over camera movement, which is critical for downstream applications related to content creation, visual effects, and 3D vision. Recently, new methods demonstrate the ability to generate videos with controllable camera poses these techniques leverage pre-trained U-Net-based diffusion models that explicitly disentangle spatial and temporal generation. Still, no existing approach enables camera control for new, transformer-based video diffusion models that process spatial and temporal information jointly. Here, we propose to tame video transformers for 3D camera control using a ControlNet-like conditioning mechanism that incorporates spatiotemporal camera embeddings based on Plucker coordinates. The approach demonstrates state-of-the-art performance for controllable video generation after fine-tuning on the RealEstate10K dataset. To the best of our knowledge, our work is the first to enable camera control for transformer-based video diffusion models.

Related papers

• VidCRAFT3: Camera, Object, and Lighting Control for Image-to-Video Generation [62.64811405314847]
  We introduce VidCRAFT3, a novel framework for precise image-to-video generation. It enables control over camera motion, object motion, and lighting direction simultaneously. Experiments on benchmark datasets demonstrate the efficacy of VidCRAFT3 in producing high-quality video content.
  arXiv  Detail & Related papers  (2025-02-11T13:11:59Z)
• Diffusion as Shader: 3D-aware Video Diffusion for Versatile Video Generation Control [73.10569113380775]
  Diffusion as Shader (DaS) is a novel approach that supports multiple video control tasks within a unified architecture. DaS leverages 3D tracking videos as control inputs, making the video diffusion process inherently 3D-aware. DaS demonstrates strong control capabilities across diverse tasks, including mesh-to-video generation, camera control, motion transfer, and object manipulation.
  arXiv  Detail & Related papers  (2025-01-07T15:01:58Z)
• AC3D: Analyzing and Improving 3D Camera Control in Video Diffusion Transformers [66.29824750770389]
  We analyze camera motion from a first principles perspective, uncovering insights that enable precise 3D camera manipulation. We compound these findings to design the Advanced 3D Camera Control (AC3D) architecture.
  arXiv  Detail & Related papers  (2024-11-27T18:49:13Z)
• Boosting Camera Motion Control for Video Diffusion Transformers [21.151900688555624]
  We show that transformer-based diffusion models (DiT) suffer from severe degradation in camera motion accuracy. To address the persistent motion degradation in DiT, we introduce Camera Motion Guidance (CMG), which boosts camera control by over 400%. Our method universally applies to both U-Net and DiT models, offering improved camera control for video generation tasks.
  arXiv  Detail & Related papers  (2024-10-14T17:58:07Z)
• Training-free Camera Control for Video Generation [19.526135830699882]
  We propose a training-free and robust solution to offer camera movement control for off-the-shelf video diffusion models. Our method does not require any supervised finetuning on camera-annotated datasets or self-supervised training via data augmentation.
  arXiv  Detail & Related papers  (2024-06-14T15:33:00Z)
• Ctrl-V: Higher Fidelity Video Generation with Bounding-Box Controlled Object Motion [8.068194154084967]
  This paper tackles a challenge of how to exert precise control over object motion for realistic video synthesis. To accomplish this, we control object movements using bounding boxes and extend this control to the renderings of 2D or 3D boxes in pixel space. Our method, Ctrl-V, leverages modified and fine-tuned Stable Video Diffusion (SVD) models to solve both trajectory and video generation.
  arXiv  Detail & Related papers  (2024-06-09T03:44:35Z)
• CamCo: Camera-Controllable 3D-Consistent Image-to-Video Generation [117.16677556874278]
  We introduce CamCo, which allows fine-grained Camera pose Control for image-to-video generation. To enhance 3D consistency in the videos produced, we integrate an epipolar attention module in each attention block. Our experiments show that CamCo significantly improves 3D consistency and camera control capabilities compared to previous models.
  arXiv  Detail & Related papers  (2024-06-04T17:27:19Z)
• Collaborative Video Diffusion: Consistent Multi-video Generation with Camera Control [70.17137528953953]
  Collaborative video diffusion (CVD) is trained on top of a state-of-the-art camera-control module for video generation. CVD generates multiple videos rendered from different camera trajectories with significantly better consistency than baselines.
  arXiv  Detail & Related papers  (2024-05-27T17:58:01Z)
• CamViG: Camera Aware Image-to-Video Generation with Multimodal Transformers [18.67069364925506]
  We propose to add virtual 3D camera controls to generative video methods by conditioning generated video on an encoding of three-dimensional camera movement. Results demonstrate that we are (1) able to successfully control the camera during video generation, starting from a single frame and a camera signal, and (2) we demonstrate the accuracy of the generated 3D camera paths using traditional computer vision methods.
  arXiv  Detail & Related papers  (2024-05-21T20:54:27Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.