PartiCam: Camera Controlled Video Generation with Reward Guidance

Amine Ouasfi; Runjia Li; Junlin Han; Eric Marchand; Philip H. S. Torr; Adnane Boukhayma

cs.CVSep 30, 2026

PartiCam: Camera Controlled Video Generation with Reward Guidance

Authors: Amine Ouasfi, Runjia Li, Junlin Han, Eric Marchand, Philip H. S. Torr, Adnane Boukhayma

Organizations: INRIA, Univ. Rennes, CNRS, IRISA · University of Oxford · Meta

Abstract

We present PartiCam, a training-free Particle filtering rooted method for improved Camera controlled video generation. Generating videos that follow a precisely specified camera trajectory remains challenging for large video diffusion models. Training-free approaches are backbone-agnostic and avoid the need to construct large camera-annotated datasets by steering pretrained models toward the desired camera motion at test time. This enables the generation of camera-controlled video data that can subsequently be used to train camera-conditioned video diffusion models. Existing sampling-based guidance approaches often suffer from unstable trajectories: they either explore too broadly and fail to respect the target camera motion or collapse early and lose visual diversity over time. We introduce a global-local refinement framework for diffusion reward guidance, enabling accurate and consistent camera control during video generation. Our method builds on Sequential Monte-Carlo (SMC) guidance, but introduces a local refinement stage based on particle filtered resampling. Experiments show large improvements in camera trajectory adherence, reduced drift, and better visual quality, without requiring model retraining.

Figures & tables

Figure 2 : One denoising step with Global SMC.

Figure 3 : Qualitative comparison of different methods on NVS from single image of static scenes. We compare to methods NVS-Solver [ 78 ] , NVS-Solver DGS [ 78 ] , MotionCtrl [ 66 ] , Text2NeRF [ 86 ] .

Methods	Overfitting	FID	ATE	RPE-T	RPE-R
SparseGS [ 72 ]	✓	369.19	–	–	–
Text2NeRF [ 86 ]	✓	187.05	2.223	0.718	0.107
Photo-NVS [ 80 ]	$\times$	193.87	7.64	1.19	1.45
3D-aware [ 71 ]	$\times$	217.19	2.836	1.258	1.662
MotionCtrl [ 66 ]	$\times$	179.24	3.851	0.705	0.835
NVS-S (DGS) [ 78 ]	$\times$	166.50	4.533	0.810	0.742

Table 1 : Quantitative comparison of different methods on NVS from single image of static scenes. For all metrics, the lower, the better.

Figure 4 : Qualitative comparison of different methods on NVS from monocular videos of dynamic scenes. We compare to methods 4DGS [ 69 ] , MotionCtrl [ 66 ] , NVS-Solver [ 78 ] , NVS-Solver DGS [ 78 ] .

Methods	Overfitting	FID	ATE	RPE-T	RPE-R
Def-Gaussian [ 46 ]	✓	115.82	1.813	0.678	0.613
4DGS [ 69 ]	✓	74.34	2.087	0.625	0.825
3D-aware [ 71 ]	$\times$	159.03	3.100	1.343	1.368
MotionCtrl [ 66 ]	$\times$	70.35	3.384	1.069	0.653
NVS-S (DGS) [ 78 ]	$\times$	37.973	2.236	0.691	0.446
NVS-S (Post) [ 78 ]	$\times$	39.86	2.308	0.725	0.400

Table 2 : Quantitative comparison of different methods on NVS from monocular videos of dynamic scenes. For all metrics, the lower, the better.

Panel of Figure 5 — Figure 5 : Ablation results of our method on Static Scenes. We use Aesthetic reward for Global SMC and Camera reward for local refinement. The number of local candidates is denoted as $N_{r}$ while the number of global particles is referred to as $K$ .

Methods	FID	ATE	RPE-T	RPE-R
NVS-S (CogVideo) [ 78 ]	41.08	0.981	0.128	0.431
Ours (CogVideo)	31.86	0.807	0.061	0.414
NVS-S (SVD) [ 78 ]	39.86	2.308	0.725	0.400
Ours (SVD)	38.26	1.912	0.521	0.489
TrajectoryCrafter [ 82 ]	29.86	0.767	0.095	0.460
Ours (TrajectoryCrafter)	30.01	0.712	0.070	0.428

Table 3 : Ablation study of our method with respect to the choice of the backbone video model for NVS from monocular videos of dynamic scenes. For all metrics, the lower, the better.

Figure 6 : Qualitative comparison in monocular video reshooting of dynamic scenes to ReCamMaster [ 3 ] .

Figure 7 : Qualitative comparison in monocular video reshooting of dynamic scenes to TrajectoryCrafter [ 82 ] . We We show a two views of the estimated camera trajectories from the videos generated with our method , TrajectoryCrafter and the Ground-truth .

Panel of Figure 8 — Figure 8 : Qualitative comparison of different methods on monocular video reshooting of dynamic scenes. We compare to methods 4DGS [ 69 ] , MotionCtrl [ 66 ] , NVS-Solver [ 78 ] , NVS-Solver DGS [ 78 ] .

Panel of Figure 9 — Figure 9 : Qualitative comparison of different methods on monocular video re-shooting of dynamic scenes. We compare to methods 4DGS [ 69 ] , MotionCtrl [ 66 ] , NVS-Solver [ 78 ] , NVS-Solver DGS [ 78 ] .

Explore similar work

Beyond Inpainting: Unleash 3D Understanding for Stable Camera-Controlled Video Generation
Jan 15, 2026Dong-Yu Chen, Yixin Guo, Shuojin Yang +2Camera Control Video Generation
BNRist, Department of Computer Science and Technology, Tsinghua University Beijing 100084, China
Track the Noise, Move the World:3D-Grounded Motion-Consistent Noise for Controllable Video Generation
Jul 2, 2026Long Vu, Tan Ngo, Animesh Karnewar +5Video Diffusion Models Controllability
Qualcomm AI Research · Trinity College Dublin, Ireland
GimbalDiffusion: Gravity-Aware Camera Control for Video Generation
Dec 9, 2025Frédéric Fortier-Chouinard, Yannick Hold-Geoffroy, Valentin Deschaintre +2Camera Control Camera Trajectories
Université Laval, Québec, Canada · Adobe, San Jose, USA

Jan 15, 2026, cs.CVJ/K move · Enter open · S save

Beyond Inpainting: Unleash 3D Understanding for Stable Camera-Controlled Video Generation

Dong-Yu Chen, Yixin Guo, Shuojin Yang, Tai-Jiang Mu+1

BNRist, Department of Computer Science and Technology, Tsinghua University Beijing 100084, China

Camera Control Video Generation Dramadirector 3D Priors 3D Representation

Open paper AskPDF arXiv

Camera control has been extensively studied in conditioned video generation; however, performing precisely altering the camera trajectories while faithfully preserving the video content remains a challenging task. The mainstream approach to achieving precise camera control is warping a 3D representation according to the target trajectory. However, such methods fail to fully leverage the 3D priors of video diffusion models (VDMs) and often fall into the Inpainting Trap, resulting in subject inconsistency and degraded generation quality. To address this problem, we propose DepthDirector, a video re-rendering framework with precise camera controllability. By leveraging the depth video from explicit 3D representation as camera-control guidance, our method can faithfully reproduce the dynamic scene of an input video under novel camera trajectories. Specifically, we design a View-Content Dual-Stream Condition mechanism that injects both the source video and the warped depth sequence rendered under the target viewpoint into the pretrained video generation model. This geometric guidance signal enables VDMs to comprehend camera movements and leverage their 3D understanding capabilities, thereby facilitating precise camera control and consistent content generation. Next, we introduce a lightweight LoRA-based video diffusion adapter to train our framework, fully preserving the knowledge priors of VDMs. Additionally, we construct a large-scale multi-camera synchronized dataset named MultiCam-WarpData using Unreal Engine 5, containing 8K videos across 1K dynamic scenes. Extensive experiments show that DepthDirector outperforms existing methods in both camera controllability and visual quality. Our code and dataset will be publicly available.

Abstract

Explore similar work

Beyond Inpainting: Unleash 3D Understanding for Stable Camera-Controlled Video Generation

Track the Noise, Move the World:3D-Grounded Motion-Consistent Noise for Controllable Video Generation

GimbalDiffusion: Gravity-Aware Camera Control for Video Generation