Camera Artist: A Multi-Agent Framework for Cinematic Language Storytelling Video Generation

Haobo Hu; Qi Mao; Yuanhang Li; Libiao Jin

cs.AIApr 10, 2026

Camera Artist: A Multi-Agent Framework for Cinematic Language Storytelling Video Generation

Authors: Haobo Hu, Qi Mao, Yuanhang Li, Libiao Jin

Organizations: School of Information and Communication Engineering, Communication University of China, Beijing, China · State Key Laboratory for Multimedia Information Processing, School of Computer Science, Peking University, Beijing, China

Abstract

We propose Camera Artist, a multi-agent framework that models a real-world filmmaking workflow to generate narrative videos with explicit cinematic language. While recent multi-agent systems have made substantial progress in automating filmmaking workflows from scripts to videos, they often lack explicit mechanisms to structure narrative progression across adjacent shots and deliberate use of cinematic language, resulting in fragmented storytelling and limited filmic quality. To address this, Camera Artist builds upon established agentic pipelines and introduces a dedicated Cinematography Shot Agent, which integrates recursive storyboard generation to strengthen shot-to-shot narrative continuity and cinematic language injection to produce more expressive, film-oriented shot designs. Extensive quantitative and qualitative results demonstrate that our approach consistently outperforms existing baselines in narrative consistency, dynamic expressiveness, and perceived film quality.

Figures & tables

Fig. 1: Comparison with multi-agent system on filmic storytelling. Existing multi-agent methods tend to exhibit fragmented narratives and weak cinematic control. In contrast, Camera Artist achieves stronger shot-to-shot coherence and richer cinematic expression, yielding more filmic storytelling.

Fig. 2: The overall framework of Camera Artist. Camera Artist operates in two stages: footage construction and shot generation . In the footage construction stage, the Director Agent expands the story outline and builds hierarchical storyboard assets at script, scene, and shot levels. In the shot generation stage, the Cinematography Shot Agent first performs recursive shot generation to ensure narrative coherence, and then injects cinematic language to refine shot descriptions. Finally, the Video Generation Agent produces shot-wise videos and stitches them into a complete long-form narrative film.

Fig. 3: Mechanism of the Cinematography Shot Agent. (a) Recursive Shots Generation (RSG): By recursively generating shots and selecting start/mid/end types, the system produces storyboards with strong narrative coherence. (b) Cinematic Language Injection (CLI): A fine-tuned LLM trained on professional cinematic language transforms original shot descriptions into film-style, cinematically expressive ones.

Fig. 4: Qualitative experimental results of single shot content. For videos with similar shot content, Camera Artist can achieve richer and more expressive cinematic language, outperforming prior multi-agent methods.

Method/Metrics	CLIP-T ( $\uparrow$ )	Subj.( $\uparrow$ )	Bg.( $\uparrow$ )	Motion( $\uparrow$ )	Dyn.( $\uparrow$ )	Aesth.( $\uparrow$ )
	Semantic Consistency	VBench Metrics
VGoT [ 21 ]	28.15	78.58	97.93	99.27	16.67	68.73
Anim-Director [ 11 ]	23.86	67.79	94.15	96.54	39.78	67.24
MovieAgent [ 18 ]	22.25	71.01	94.52	98.00	76.27	65.63
Ours	29.61	79.54	96.26	99.32	80.00	69.51

TABLE I: Quantitative comparison using VBench and CLIP-based semantic consistency. Best and second-best results are highlighted in blue and green , respectively.

Method/Metrics	GPT-4o	Qwen3	Gemini	Avg.	GPT-4o	Qwen3	Gemini	Avg.	GPT-4o	Qwen3	Gemini	Avg.	GPT-4o	Qwen3	Gemini	Avg.
	Script Cons. ( $\uparrow$ )				Cam. Cons. ( $\uparrow$ )				Video Qual. ( $\uparrow$ )				Real. Sim. ( $\uparrow$ )
VGoT [ 21 ]	3.33	3.00	2.17	2.83	1.83	1.00	1.17	1.33	4.67	4.83	2.67	4.06	4.17	3.67	2.67	3.50
Anim-Director [ 11 ]	3.60	2.50	2.83	2.98	2.40	1.00	3.50	2.30	3.33	3.17	2.67	3.06	2.00	1.83	1.00	1.61
MovieAgent [ 18 ]	2.10	1.30	3.17	2.19	1.70	1.00	3.38	2.03	4.10	4.30	4.00	4.13	2.90	2.70	3.80	3.13
Ours	4.50	4.00	3.20	3.90	3.25	3.90	3.50	3.55	4.86	4.78	4.50	4.71	4.00	4.56	3.50	4.02

TABLE II: Multi-VLM evaluation across narrative and cinematic dimensions. Best and second-best results are highlighted in blue and green , respectively.

Method	CLIP-T	Subj.( $\uparrow$ )	Bg.( $\uparrow$ )	Motion( $\uparrow$ )	Dyn.( $\uparrow$ )	Aesth.( $\uparrow$ )	Script Cons.( $\uparrow$ )	Cam. Cons.( $\uparrow$ )	Video Qual.( $\uparrow$ )	Real. Sim.( $\uparrow$ )
	Semantic	VBench Metrics					LLM-based Evaluation
w/o RSG	28.22	74.69	93.93	99.04	78.67	67.45	3.55	3.36	4.45	3.91
w/o CLI	29.27	73.49	94.33	98.97	74.25	67.10	3.60	2.83	4.00	3.67
Camera Artist	29.61	79.54	96.26	99.32	80.00	69.51	3.90	3.55	4.71	4.02

TABLE III: Quantitative results of the ablation study on recursive storyboard generation and cinematic language injection. Best and second-best results are highlighted in blue and green , respectively.

Fig. 5: Qualitative comparison of inter-shot narrative coherence. Camera Artist conditions each shot on preceding shot and scene information, producing shot content that is narratively coherent in both text and visual realization.

Fig. 6: User study comparison on four subjective metrics. Results of VGoT [ 21 ] , Anim-Director [ 11 ] , MovieAgent [ 18 ] , and our method on Script Consistency, Camera-Movement Consistency, Video Quality, and Real-Movie Similarity. Our method achieves the highest scores across all metrics.

Fig. 7: Ablation study on RSG and CLI. RSG preserves coherent shot-to-shot narrative flow, while CLI enhances cinematic expressiveness through deliberate camera motion and lighting; removing either results in fragmented storytelling or visually static shots.

Appendix figures & tables7 assets

Supplementary material from the paper’s appendix.

Appendix

Fig. 8: Camera Artist workflow visualization. Given a user-provided story outline, Camera Artist decomposes the narrative into structured scene plots and character assets via the Director Agent , refines them into coherent shot-level descriptions with explicit cinematic language using the Cinematography Shot Agent , and finally renders corresponding visual clips through the Video Generation Agent . The collaboration among agents enables automated long-form video generation with coherent narrative progression and expressive cinematic shot design.

Fig. 9: An example of pipeline for cinematic language LoRA fine-tuning. Ordinary captions are produced by a VLM from raw video, while ShotBench [ 12 ] provides shot-level cinematic annotations. A LoRA-tuned LLM learns to transform ordinary captions into cinematic shot descriptions with explicit cinematic language, which are later used for cinematic language injection during inference.

Fig. 10: The CoT Description of Camera Artist. (a) The CoT of Director Agent, which is mainly responsible for the expansion of script content and scene splitting.(b) The CoT of Cinematography Shot Agent, which is mainly responsible for the recursive generation of storyboard content and the introduction of shot language.

Fig. 11: The CoT prompting of VLM-based evaluation. The CoT is mainly responsible for the recursive generation of storyboard content and the introduction of shot language.

Fig. 12: Qualitative comparison with baseline methoda. (a) Camera Artist generates a final wide shot with high-angle composition and slow pull-back movement, delivering stronger cinematic atmosphere and expressive visual storytelling. (b) Baselines introduce irrelevant characters or exhibit abrupt narrative jumps in two-shot sequences, while Camera Artist maintains both character/scene consistency and coherent event progression.

Fig. 13: Reference-free storytelling video generation. Given only a textual story outline (no character reference images), Camera Artist automatically constructs scenes, characters, and shot sequences, producing a long-form narrative video with coherent story progression and cinematic visual expression.

Fig. 14: Additional qualitative results. Scene-level keyframes together with the corresponding footage are presented, illustrating coherent long-range storytelling, consistent character depiction, and film-style visual expression.

Explore similar work

Sep 10, 2026

CamPilot: A Multi-Agent Cinematic Assistant for Camera-Controlled Movie Generation

The integration of large language models (LLMs) into video generation has enabled rapid text-to-video creation and improved visual quality. However, it still falls short of professional filmmaking, where cinematographic language is less refined than human-crafted camera work and multi-shot continuity remains challenging. To address these limitations, we introduce CamPilot, a multi-agent framework that integrates cinematographic planning and camera-work control to produce more coherent, logically structured, and human-aesthetic movies. CamPilot adopts a GRPO-based learning paradigm to learn camera work planning from 14K real-world professional movies, internalizing motion patterns and composition principles that support reasoning over shooting techniques (e.g., camera angle, motion, and focal behavior) and cross-shot relationships for controllable camera-viewpoint generation. Multiple agents further collaborate and evolve to improve overall output quality. To support this work and further studies in this domain, we establish CamEval, a benchmark for evaluating camera work quality and cinematic engagement. Empirical results show that CamPilot outperforms state-of-the-art text-to-movie generation methods on cinematographic control and quality, highlighting the impact of professional camera design on movie generation.

Yang Wu, Stefano Petrangeli, Ishita Dasgupta +1

Worcester Polytechnic Institute, Worcester, MA, USA · Adobe Research, San Jose, CA, USA

Camera-Controlled Video Generation Multi-Shot Video Generation Video Generation Evaluation

Ask PDF

Sep 7, 2026

Better Call CineCrew: Consistent Ultra-Long Narrative-to-Film Generation

Long-form narrative-to-film generation requires shot-level controllability and cross-clip consistency in both visual identity and character behavior-requirements that remain difficult to satisfy with current prompt-based workflows. A core reason existing workflows remain brittle is the lack of a structured intermediate layer between scripts and video models, especially when screenplays are underspecified at key cinematic decision points. We introduce a structured orchestration layer for film-oriented script-to-video generation, implemented as a multi-agent framework that operates between scripts and off-the-shelf video generators. The layer is centered on FilmDSL, a film-oriented domain-specific language that makes cinematic constraints explicit, including shot and camera directives, asset and continuity requirements, and persona cues, so that agents coordinate through a shared structured specification for planning, generation, critique, and repair. Specifically, a generation agent constructs asset packs and storyboard keyframes that anchor composition before clip-by-clip synthesis, while a critic agent produces structured QA signals and triggers targeted refinement without retraining the base model. Experiments on TV-style segments show improved controllability and consistency over text-only and reference-only baselines.

Jiaben Chen, Sixun Dong, Qinhong Zhou +4

University of Massachusetts Amherst, Amherst, MA 01003, USA · Massachusetts Institute of Technology, Cambridge, MA 02139, USA

Text-to-Video Generation Long Video Generation Agentic Video Generation Identity-Preserving Video Generation Multi-Shot Video Generation Temporal Consistency in Video Generation

Ask PDF

Aug 30, 2026

CineForge: Self-Improving Agents for Long-Horizon Video Generation

Long-horizon story-driven video generation requires a production agent to coordinate narrative decomposition, state tracking, shot design, prompt construction, rendering, and revision across interdependent scenes. Existing adaptive video systems primarily refine requests or reusable skills, leaving recurring production failures disconnected from persistent, stage-targeted improvements across stories. We introduce CineForge, a self-evolving video-production agent framework that couples CineForge-Produce for video generation with CineForge-Evolve for cross-story policy evolution. CineForge-Produce organizes each source story into typed narrative, character, spatial, and cinematic states, uses them to coordinate asset and clip generation, and records the process as a canonical production trajectory. CineForge-Evolve applies Case-to-Pattern-to-Policy Evolution (CPPE) to review trajectory evidence, consolidate recurrent findings into bounded stage-local patches, and deploy validated updates through structural replay and confidence-controlled paired evaluation. To measure complete story realization, we introduce CineScope, which combines a 100-script CineScope-Data suite with a human-aligned, multiscale CineScope-Metric spanning causal state, directorial orchestration, pacing and resource allocation, and character arc. Across CineScope-Data and two public benchmarks, the evolved CineForge policy improves CineScope-Metric from 4.024 to 4.380, outperforms three long-video baselines with consistent gains under ScriptAgent, and reduces review LLM calls by 37.0% on new stories. These results establish production trajectories as actionable experience for video agents that improve cumulatively across long-form storytelling tasks.

Junxiang Liu, Lin Wang, Haiyu Shi +10

Tongji University, Shanghai, China · Kuaishou Technology, Beijing, China · Fudan University, Shanghai, China

Long-Horizon Video Generation Agentic Video Generation Multi-Shot Video Generation Self-Improving Agents Long-Horizon LLM Agents Video Generation Evaluation

Ask PDF