Researchers at the Technion-Israel Institute of Technology have developed a technology that allows users to control movement in AI-generated videos using simple mouse gestures, without requiring large computing resources or retraining on massive video datasets.
The system, called Time to Move, or TTM, was developed by Dr. Or Litany of the Henry and Marilyn Taub Faculty of Computer Science, together with Prof. Ron Kimmel and students Asaf Singer, Noam Rotstein and Amir Mann.
Technion tool lets users create AI videos by mouse
(Video: Technion)
Litany presented the research last month at the International Conference on Learning Representations, or ICLR 2026, in Brazil. The conference is considered one of the leading global gatherings in deep learning and artificial intelligence.
The technology is designed to address one of the key limitations of AI video generation: the difficulty of precisely controlling how objects and characters move over time. “Our development solves one of the main limitations of AI-based video generation: the difficulty of precisely controlling the movement of objects and characters over time,” Litany said.
He said TTM can be integrated as a plug-in into existing video models and does not require retraining. Unlike earlier approaches that require model-specific adaptation and significant computing power, the Technion system operates without additional computational cost, he said.
“In doing so, it helps democratize AI video creation by expanding access beyond giant companies such as Google and Meta,” Litany said.
Dr. Or Litany Photo: Technion1 View gallery
Demonstration of the new technology compared with existing technologies. In each image pair, the left side shows current capabilities, while the right side demonstrates the capabilities of TTM
(Illustration: Technion)
The key innovation behind the technology is a method called dual-clock denoising, which refines motion while balancing the user’s intended movement with natural-looking video results.
Experiments conducted by the researchers showed that TTM matched training-based methods and outperformed them in motion accuracy and realism, according to the Technion. The system also allows users to edit the appearance of objects and add new objects to a scene, capabilities not offered by some earlier trained methods.
Researchers said the technology represents a step toward more intuitive and controllable tools for generative video.
Litany joined the Technion’s computer science faculty as a senior lecturer in 2023 after being selected as an Azrieli Faculty Fellow and a Taub Fellow. He previously completed postdoctoral fellowships at Stanford University and FAIR at Meta and has worked on computer vision technologies.