In-Video Instructions: Visual Signals as Generative Control

TL;DR

This paper introduces In-Video Instruction, a method for controlling video generation by embedding visual signals like text and arrows directly into frames, enabling explicit and spatially-aware user guidance.

Contribution

It proposes a novel visual signal-based control paradigm for video generation, allowing explicit, spatially-aware instructions embedded directly in frames.

Findings

Video models reliably interpret embedded visual instructions

Effective control demonstrated in complex multi-object scenarios

Compatible with multiple state-of-the-art generators

Abstract

Large-scale video generative models have recently demonstrated strong visual capabilities, enabling the prediction of future frames that adhere to the logical and physical cues in the current observation. In this work, we investigate whether such capabilities can be harnessed for controllable image-to-video generation by interpreting visual signals embedded within the frames as instructions, a paradigm we term In-Video Instruction. In contrast to prompt-based control, which provides textual descriptions that are inherently global and coarse, In-Video Instruction encodes user guidance directly into the visual domain through elements such as overlaid text, arrows, or trajectories. This enables explicit, spatial-aware, and unambiguous correspondences between visual subjects and their intended actions by assigning distinct instructions to different objects. Extensive experiments on three state-of-the-art generators, including Veo 3.1, Kling 2.5, and Wan 2.2, show that video models can reliably interpret and execute such visually embedded instructions, particularly in complex multi-object scenarios.