High-Accuracy Stereo Depth Maps Using Structured Light

Daniel Scharstein and Richard Szeliski
In IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2003), volume 1, pages 195-202, Madison, WI, June 2003.

The paper (pdf, 1.2M)

Abstract

Recent progress in stereo algorithm performance is quickly outpacing the ability of existing stereo data sets to discriminate among the best-performing algorithms, motivating the need for more challenging scenes with accurate ground truth information. This paper describes a method for acquiring high-complexity stereo image pairs with pixel-accurate correspondence information using structured light. Unlike traditional range-sensing approaches, our method does not require the calibration of the light sources and yields registered disparity maps between all pairs of cameras and illumination projectors. We present new stereo data sets acquired with our method and demonstrate their suitability for stereo algorithm evaluation. Our results are available at http://www.middlebury.edu/stereo.

On-line figures

Figure 1: Experimental setup

A 10-bit horizontal Gray-code sequence.

Figure 2: Examples of thresholded Gray-code images.

Sine wave patterns

Figure 4: Computed (u,v) coordinates

Figure 5: The two image pairs: Cones and Teddy

Figure 6: Left view of Cones:
(a) scene under illumination
(b) view disparities
(c) illumination disparities
(d) final (combined) disparity map

Figure 7: Left view of Teddy:
(a) scene under the two illuminations
(b) view disparities
(c) illumination disparities
(d) final (combined) disparity map

Figure 8: Stereo results on cropped and downsampled images
Table 1 (updated): Performance of stereo algorithms