home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
vis-ftp.cs.umass.edu
/
vis-ftp.cs.umass.edu.tar
/
vis-ftp.cs.umass.edu
/
pub
/
Motion
/
README
< prev
Wrap
Text File
|
1995-03-22
|
9KB
|
317 lines
Motion and Stereo sequences available from University of
Massachusetts, Amherst.
------------------------------------------------------------------
Sequence #1:
Sequence name: UMASS ROCKET FIELD sequence
Organization: University of Massachusetts, Amherst
Project:: Visions
Sponsor(s): ARPA
Environment: Outdoor; with cones, boxes, poles, buildings, mountains
Task: Depth determination
Ground truth?: Yes
Brief description: The ground truth includes the depths of scene
points obtained using theodolites as well as data on the motion of the
vehicle and appropriate coordinate transformations. Images collected
by Umass with the ALV.
Reference(s): R. Dutta, R. Manmatha, L. R. Williams and E. M. Riseman,
"A Data Set for Quantitative Motion Analysis," Proceedings of the IEEE
Computer Society Conference on Computer Vision and Pattern
Recognition, June 1989.
FTP: vis-ftp.cs.umass.edu:/Motion/motion_data
HTTP: http://vis-www.cs.umass.edu/vislib/Motion/motion_data/directory.html
------------------------------------------------------------------
Sequence #2:
Sequence name: UMASS INDOOR CONE sequence
Organization: University of Massachusetts, Amherst
Project: Visions
Sponsor(s): ARPA
Environment: Indoor; cones in a hallway
Task: Depth determination
Ground truth?: Yes
Brief description: The ground truth includes the depths of cones and
other objects.
Reference(s): H. S. Sawhney and A. R. Hanson, "Trackability as a Cue
for Potential Obstacle Identification and 3D Description," IJCV, Vol,
11, No. 3, 1993, pp. 237--265.
FTP: vis-ftp.cs.umass.edu:/Motion/hallway_cones
HTTP: http://vis-www.cs.umass.edu/vislib/Motion/hallway_cones/images.html
------------------------------------------------------------------
Sequence #3:
Sequence name: UMASS "CMU" sequence
Organization: University of Massachusetts, Amherst
Project: Visions
Sponsor(s): ARPA
Environment: Outdoor; with cones, cans, trees, buildings.
Task: Depth determination
Ground truth? : Yes
Brief description: The ground truth includes the depths of cones
and cans. Images collected by Umass using the CMU Navlab.
Reference(s): R. Dutta, "Depth from Motion and Stereo: Parallel and
Sequential Algorithms, Robustness and Lower Bounds", Ph.D.
dissertation, Computer Science Dept., University of Massachusetts at
Amherst, 1994.
R. Dutta, R. Manmatha, E. Riseman, M. Snyder,
"Issues in Extracting Motion Parameters and Depth from Approximate
Translational Motion," DARPA Image Understanding Workshop, April 1988,
pp. 945-960.
FTP: vis-ftp.cs.umass.edu:/Motion/motion_data/cones_can_sequence
HTTP: http://vis-www.cs.umass.edu/vislib/Motion/motion_data/cones_can_sequence/images.html
------------------------------------------------------------------
Sequence #4:
Sequence name: UMASS PUMA sequence
Organization: University of Massachusetts, Amherst
Project: Visions
Sponsor(s): ARPA
Environment: Robotics lab at UMASS plastered with posters on the walls
and floor.
Task: Pose determination, Structure from motion
Ground truth?: Yes, 3D location of scene points available.
Brief description: The sequence was taken by connecting a camera to
the end of a puma robot arm. The robot arm was then rotated for 120
deg. and 30 image frames were taken each frame being taken
approximately after every 4 deg.. The radius of the robot arm was
approx. 1.8 feet.
Reference(s): Kumar, R. and Hanson, A.R. (1990). ``Sensitivity of Pose
Refinement to Accurate Estimation of Camera Parameters''. Proceedings of
IEEE International Conference on Computer Vision, Osaka, Japan,
December 1990.
Kumar, R. and Hanson, A.R. (1992). ``Model Extension and
Refinement using Pose Recovery Techniques''. Journal of Robotic Systems,
Sept. 1992.
FTP: vis-ftp.cs.umass.edu:/Motion/puma2
HTTP: http://vis-www.cs.umass.edu/vislib/Motion/puma2/images.html
------------------------------------------------------------------
Sequence #5:
Sequence name: UMASS BOX sequence
Organization: University of Massachusetts, Amherst
Project: Visions
Sponsor(s): ARPA
Environment: Images of a Checkered Box Rotating about central axis every
four degrees.
Task: Pose determination; depth and motion determination using a
rotational model, as well as for two-frame and multi-frame SFM.
Ground truth?: Yes, 3D location of scene points available.
Brief description: Distinctive marks for pose estimation.
Reference(s): R. Kumar and A. R. Hanson, ``Sensitivity of Pose
Refinement to Accurate Estimation of Camera Parameters''. Proceedings of
IEEE International Conference on Computer Vision, Osaka, Japan,
December 1990.
H.S. Sawhney, J. Oliensis and A. R. Hanson, ``Image
description and 3-D reconstruction from image trajectories of rotational
motion''. IEEE Transactions on Pattern Analysis and Machine Intelligence,
Vol. 15, No. 9, Sep. 1993.
H. S. Sawhney and A. R. Hanson, "Comparative Results
of Some Motion Algorithms on Real Image Sequences," DARPA IUW, 1990.
FTP: vis-ftp.cs.umass.edu:/Motion/box
HTTP: http://vis-www.cs.umass.edu/vislib/Motion/box/images.html
------------------------------------------------------------------
Sequence #6:
Sequence name: UMASS MOTION-STEREO sequence
Organization: University of Massachusetts, Amherst
Project: Visions
Sponsor(s): ARPA
Environment: Indoor
Task: Depth determination from stereo-motion
Ground truth?: Yes
Brief description: Indoor run of four steps with a Denning robot
vehicle under approximately translational motion of about 1.9 feet at
each step. The baseline employed is 20 inches and the range of depth
about 30 feet.
Reference(s): Poornima Balasubramanyam and M. A. Snyder, "A
Computational Model for Binocular Motion Processing," Proceedings of
the ARPA Image Understanding Workshop, 1992.
FTP: vis-ftp.cs.umass.edu:/Motion/stereo_motion_1
HTTP: http://vis-www.cs.umass.edu/vislib/Motion/stereo_motion_1/images.html
------------------------------------------------------------------
Sequence #7:
Organization: University of Massachusetts, Amherst (Courtesy of Dr.
Larry Matthies)
Project: Visions
Sponsor(s): ARPA
Environment: Indoor
Task: Depth determination from stereo-motion
Ground truth?: Yes
Brief description: Indoor run with the Neptune vehicle at CMU. It
consists of 69 steps straight forward, where each step is 3.8 inches.
The images are calibrated with an image warping algorithm developed by
Moravec.
Reference(s): Poornima Balasubramanyam and M. A. Snyder, "A
Computational Model for Binocular Motion Processing," Proceedings of
the ARPA Image Understanding Workshop, 1992.
FTP: vis-ftp.cs.umass.edu:/Motion/may21
HTTP: http://vis-www.cs.umass.edu/vislib/Motion/may21/images.html
------------------------------------------------------------------
Sequence #8:
Sequence name: UMASS A211 sequence
Organization: University of Massachusetts, Amherst
Project: Visions
Sponsor(s): ARPA
Environment: Indoor
Task: Recovery of shallow structure
Ground truth?: Yes
Brief description: Indoor image sequence of the vision lab at Umass
Reference(s): H. S. Sawhney and A. R. Hanson, "Trackability as a Cue
for Potential Obstacle Identification and 3D Description," IJCV, Vol,
11, No. 3, 1993, pp. 237--265.
FTP: vis-ftp.cs.umass.edu:/Motion/a211
HTTP: http://vis-www.cs.umass.edu/vislib/Motion/a211/images.html
------------------------------------------------------------------
Sequence #9:
Sequence name: UMASS A211 sequence
Organization: University of Massachusetts, Amherst
Project: Visions
Sponsor(s): ARPA
Environment: Indoor
Task: Recovery of shallow structure
Ground truth?: Yes
Brief description: Indoor image sequences in the computer science dept., UMass.
Reference(s): H. S. Sawhney and A. R. Hanson, "Trackability as a Cue
for Potential Obstacle Identification and 3D Description," IJCV, Vol,
11, No. 3, 1993, pp. 237--265.
FTP: vis-ftp.cs.umass.edu:/Motion/coins
HTTP: http://vis-www.cs.umass.edu/vislib/Motion/coins/images.html
------------------------------------------------------------------
Sequence #10:
Sequence name: UMASS MOTION-STEREO stadium sequence
Organization: University of Massachusetts, Amherst
Project: Visions
Sponsor(s): ARPA
Environment: Outdoor University Stadium
Task: Obstacle Detection
Ground truth?: Yes
Brief description: Outdoor University Stadium Motion-stereo sequence taken
by MPL vehicle, UMass.
Reference(s): Z. Zhang, R. Weiss and A. R. Hanson, Qualitative Obstacle
Detection, CVPR, 1994.
FTP: vis-ftp.cs.umass.edu:/Miscellaneous/stadium
HTTP: http://vis-www.cs.umass.edu/vislib/Miscellaneous/stadium/images.html
------------------------------------------------------------------