ACTS is an automatic camera tracking system which can efficiently and stably recover camera motion as well as 3D scene strucutre from videos and film sequences. It is for non-commercial research and educational use ONLY. Not for reproduction, distribution or commercial use. If you use this executable for your academic publication, please acknowledge our work. This program is tested on Windows XP, Server 2003, but is still not guaranteed to be bug-free and work properly with all versions of Windows. You are welcome to report any suggestions or bugs. We will actively update the program. Please email Guofeng Zhang if you have any questions.

Structure-from-Motion is a cornerstone for many other computer vision tasks, such as 3D reconstruction, video-based rendering and video editing. In the future, we will actively update ACTS and add more and more advanced or extra functions along with our published papers, such as non-consecutive feature tracking, video stabilization, stereoscopic video synthesis, and depth video recovery.

This is a maintenance release with update. The following major new features were added:

1) Add a new feature tracking method "ENFT: Efficient Non-consecutive Feature Tracking". This method implements the consecutive point tracking step introduced in this paper: Guofeng Zhang, Zilong Dong, Jiaya Jia, Tien-Tsin Wong, and Hujun Bao. Efficient Non-Consecutive Feature Tracking for Structure-from-Motion. European Conference on Computer Vision (ECCV), 2010.
2) Add a function for automatically optimizing the threshold of minimum track length. Menu: Actions/Key Frames, click "Keyframe" to open keyframe selection dialog.

Please refer to Changelog for more update details.

• Structure-from-Motion:
  • Guofeng Zhang, Xueying Qin, Wei Hua, Tien-Tsin Wong, Pheng-Ann Heng and Hujun Bao. Robust Metric Reconstruction from Challenging Video Sequences. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2007. [paper, video]
  • Guofeng Zhang, Zilong Dong, Jiaya Jia, Tien-Tsin Wong, and Hujun Bao. Efficient Non-Consecutive Feature Tracking for Structure-from-Motion. European Conference on Computer Vision (ECCV), 2010.[paper, video]
• Video Stabilization:
  • Guofeng Zhang, Wei Hua, Xueying Qin, Yuanlong Shao, and Hujun Bao. Video Stabilization Based on a 3D Perspective Camera Model. The Visual Computer, 25(11): 997-1008, 2009. [paper]

• Constant focal length: Road Sequence;    Varying focal length: Syn. Campus Sequence, Garden Sequence, Building Sequence
• A simple user manual (pdf, video) (last updated on Aug. 29, 2010)
• Click here to download the ACTS Executable.
  • If you did not receive the download URL email, please directly email Guofeng Zhang for it.

• Guofeng Zhang: Design the system framework. Develop and maintain the kernel.
• Zilong Dong: Design and implement most of the UI. Also implement some auxiliary code/tools.
• Hanqing Jiang: Implement the interactive 3D viewer.
• Yuanlong Shao : Implement some part of the UI, including timeline and play control.

• Depth Video Recovery
• Refilming with Depth-Inferred Videos
• Realtime Camera Tracking