ObjCtrl-2.5D: Training-free Object Control with Camera Poses

TL;DR

ObjCtrl-2.5D introduces a training-free method for precise object control in image-to-video generation by modeling 3D trajectories as camera poses, enabling complex and diverse object motions without additional training.

Contribution

It proposes a novel training-free approach that uses 3D camera pose sequences for object control, extending 2D trajectories with depth, and adapts existing models for local object motion control.

Findings

Significantly improves object control accuracy over existing training-free methods.

Enables complex object motions like rotation with diverse control capabilities.

Demonstrates effectiveness through extensive experiments.

Abstract

This study aims to achieve more precise and versatile object control in image-to-video (I2V) generation. Current methods typically represent the spatial movement of target objects with 2D trajectories, which often fail to capture user intention and frequently produce unnatural results. To enhance control, we present ObjCtrl-2.5D, a training-free object control approach that uses a 3D trajectory, extended from a 2D trajectory with depth information, as a control signal. By modeling object movement as camera movement, ObjCtrl-2.5D represents the 3D trajectory as a sequence of camera poses, enabling object motion control using an existing camera motion control I2V generation model (CMC-I2V) without training. To adapt the CMC-I2V model originally designed for global motion control to handle local object motion, we introduce a module to isolate the target object from the background, enabling independent local control. In addition, we devise an effective way to achieve more accurate object control by sharing low-frequency warped latent within the object's region across frames. Extensive experiments demonstrate that ObjCtrl-2.5D significantly improves object control accuracy compared to training-free methods and offers more diverse control capabilities than training-based approaches using 2D trajectories, enabling complex effects like object rotation. Code and results are available at https://wzhouxiff.github.io/projects/ObjCtrl-2.5D/.