Videoinu

Videoinu is a web-based video production platform designed for creators who prefer a scene-driven, narrative-first approach to filmmaking rather than relying on abstract text prompts or jumping between multiple software tools. The core workflow centers on writing individual scenes in plain language, after which an AI agent automatically handles the heavy lifting of generating consistent visuals, key frames, character identities, and final video assembly. To begin, a user types a scene description — for example, “Stage a cinematic duel — Zhao Zilong versus Lu Bu, wuxia photorealism.” The AI agent interprets this input, drafts corresponding character prompts, renders each character with distinct visual attributes (e.g., Zhao Zilong with a lightning spear, Lu Bu with a flaming halberd), and locks their identity across all future frames. This identity consistency is a key feature: once a character is introduced in a scene, later shots referencing the same character automatically retain the same appearance, lighting, and costume, eliminating the need for manual re-prompting or frame-by-frame retouching. The platform then generates key frames for each scene. Users can view these frames and adjust them through a Canvas visual editor, which provides tools for composition, color grading, overlay effects, and timing control. The agent also composes the final shot by combining the generated key frames, transitions, and audio layers into a seamless video sequence. Throughout the process, the user’s role shifts from manual creator to director: they watch the results, steer the narrative by rewriting scene text, and approve or reject generated outputs before final export. Videoinu supports over 30 built-in AI models for effects, style transfer, character generation, and background sy