Motion Blur/Motion Blur Free Dataset
Introduction
The Motion Blur/Motion Blur Free Dataset consists of 16 stereo videos showing a non-static scene. The cameras were assembled on a 3-D mirror rig and not moved during the recording. They captured the scene with different exposures to obtain blurry and sharp frames of objects moving in front of the camera. Some objects were even printed, such that their geomery and texture are known.
Acquisition
Setup: Two cameras were assembled on a 3-D mirror rig to capture the scene from nearly the same point of view:
Both cameras captured the scene with a resolution of 4096x2160 at 25fps, except for the videos Juggler3 and Juggler4 with 50fps. The exposure of the camera capturing blurry frames was set to 150.50deg in nearly all setups and the exposure of the other camera to 18.00deg. Exceptions are the Juggler videos (exposure times are written above the videos). The cameras were synchronized such that their exposure intervals end at the same time. To ensure a correct synchronization of the cameras, we recorded a Camera Synchronization Running Light. A ND-Filter was used on the camera captuering the scene with the longer exposure to obtain the same brightness in both videos.
Printed 3-D Objects:- Granite head of Amenemhat III
- A Bronze Age Sussex Loop Bracelet
- Stanford Bunny (own texture)
- Stanford Dragon (untextured)
Beside these printed 3-D objects we captured also some people juggling with different balls, calibration bodies, a Camera Synchronization Running Light and planar objects (textured aluminium boards and a Color Checkerboard).
Preview
| Granite head of Amenemhat III | A Bronze Age Sussex Loop Bracelet | |
|---|---|---|
| long exposure |
Download Preview: Amenemhat_Blurry.mp4 |
Download Preview: Bracelet_Blurry.mp4 |
| short exposure |
Download Preview: Amenemhat_Sharp.mp4 |
Download Preview: Bracelet_Sharp.mp4 |
| Stanford Bunny | Stanford Dragon | |
| long exposure |
Download Preview: Bunny_Blurry.mp4 |
Download Preview: Dragon_Blurry.mp4 |
| short exposure |
Download Preview: Bunny_Sharp.mp4 |
Download Preview: Dragon_Sharp.mp4 |
| Color Checkerboard | Camera Synchronization Running Light | |
| long exposure |
Download Preview: ColorChecker_Blurry.mp4 |
Download Preview: RunningLights_Blurry.mp4 |
| short exposure |
Download Preview: ColorChecker_Sharp.mp4 |
Download Preview: RunningLights_Sharp.mp4 |
| Calibration Body I | Calibration Body II | |
| long exposure |
Download Preview: Calib1_Blurry.mp4 |
Download Preview: Calib2_Blurry.mp4 |
| short exposure |
Download Preview: Calib1_Sharp.mp4 |
Download Preview: Calib2_Sharp.mp4 |
| Aluminum Plate I | Aluminum Plate II | |
| long exposure |
Download Preview: AluPlate1_Blurry.mp4 |
Download Preview: AluPlate2_Blurry.mp4 |
| short exposure |
Download Preview: AluPlate1_Sharp.mp4 |
Download Preview: AluPlate2_Sharp.mp4 |
| Aluminum Plate III | Aluminum Plate IV | |
| long exposure |
Download Preview: AluPlate3_Blurry.mp4 |
Download Preview: AluPlate4_Blurry.mp4 |
| short exposure |
Download Preview: AluPlate3_Sharp.mp4 |
Download Preview: AluPlate4_Sharp.mp4 |
| Juggler I (Shutter angle: 72.5deg/9.0deg) |
Juggler II (Shutter angle: 72.5deg/9.0deg) |
|
| long exposure |
Download Preview: Juggler1_Blurry.mp4 |
Download Preview: Juggler2_Blurry.mp4 |
| short exposure |
Download Preview: Juggler1_Sharp.mp4 |
Download Preview: Juggler2_Sharp.mp4 |
| Juggler III (Shutter angle: 144.5deg/18.0deg) |
Juggler IV (Shutter angle: 144.5deg/18.0deg) |
|
| long exposure |
Download Preview: Juggler3_Blurry.mp4 |
Download Preview: Juggler4_Blurry.mp4 |
| short exposure |
Download Preview: Juggler3_Sharp.mp4 |
Download Preview: Juggler4_Sharp.mp4 |
Requesting the data
Please send an e-mail to C. Seibold (clemens.seibold@hhi.fraunhofer.de) to request access to the database. All submitted papers or any publicly available text using the Motion Blur/Motion Blur Free Dataset must cite the following paper:
Related publication
Seibold C., Hilsmann A. and Eisert P., Model-Based Motion Blur Estimation for the Improvement of Motion Tracking, Computer Vision and Image Understanding, vol. 160, pp. 45-56, July 2017.