Robustness evaluation of stereo algorithms on long stereo sequences

Morales, Sandino; Vaudrey, Tobi; Klette, Reinhard

doi:10.1109/ivs.2009.5164302

Cited by 25 publications

(19 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As expected with noise-free stereo pairs, the SGM results were 'very close' to the ground truth, letting almost no room for improvement. All the filtered results were slightly worse than the raw stereo results, as already reported in [Morales et al 2009]. However, this previous study reported that using either spatial or temporal post-processing leads to improvements when using noisy sequences which resemble real-world data.…”

Section: Experiments and Resultssupporting

confidence: 52%

“…The basic idea of our approach is to incorporate disparity information contained in the spatial domain, as reported in [Morales et al 2009], into the Kalman filter approaches presented in [Badino et al 2007,Vaudrey et al 2008a. These previous Kalman filter approaches were designed to handle different kinds of moving objects.…”

Section: Approachmentioning

confidence: 99%

“…For example, in [Morales et al 2009], an 'alpha-blending' of disparity values calculated at current and previous image frames lead to an improved performance for the currently measured values. This model did not yet consider the egomotion induced by the ego-vehicle (i.e.…”

Section: Introductionmentioning

confidence: 99%

“…using data from the spatial domain). [Morales et al 2009] used the same alpha-blending approach for modifying the disparity value at a pixel by using disparity values of the 'north' and 'south' neighbours. Again, no ego-or independent motion was taken into account.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Spatio-Temporal Stereo Disparity Integration

Morales

Klette

2011

Computer Analysis of Images and Patterns

Self Cite

View full text Add to dashboard Cite

Abstract. Using image sequences as input for vision-based algorithms allows the possibility of merging information from previous images into the analysis of the current image. In the context of video-based driver assistance systems, such temporal analysis can lead to the improvement of depth estimation of visible objects. This paper presents a Kalman filter-based approach that focuses on the reduction of uncertainty in disparity maps of image sequences. For each pixel in the current disparity map, we incorporate disparity data from neighbourhoods of corresponding pixels in the immediate previous and the current image frame. Similar approaches have been considered before that also use disparity information from previous images, but without complementing the analysis with data from neighbouring pixels.

show abstract

Section: Experiments and Resultssupporting

confidence: 52%

Section: Approachmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Spatio-Temporal Stereo Disparity Integration

Morales

Klette

2011

Computer Analysis of Images and Patterns

Self Cite

View full text Add to dashboard Cite

show abstract

“…Another evaluation example using several sensors of the same type was published in [28]. In this approach various common stereo matching algorithms were evaluated using three cameras.…”

Section: Ground-truthingmentioning

confidence: 99%

An Evaluation Framework for Stereo-Based Driver Assistance

Schneider

Gehrig

Pfeiffer

et al. 2012

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. The accuracy of stereo algorithms or optical flow methods is commonly assessed by comparing the results against the Middlebury database. However, equivalent data for automotive or robotics applications rarely exist as they are difficult to obtain. As our main contribution, we introduce an evaluation framework tailored for stereo-based driver assistance able to deliver excellent performance measures while circumventing manual label effort. Within this framework one can combine several ways of ground-truthing, different comparison metrics, and use large image databases. Using our framework we show examples on several types of groundtruthing techniques: implicit ground truthing (e.g. sequence recorded without a crash occurred), robotic vehicles with high precision sensors, and to a small extent, manual labeling. To show the effectiveness of our evaluation framework we compare three different stereo algorithms on pixel and object level. In more detail we evaluate an intermediate representation called the Stixel World. Besides evaluating the accuracy of the Stixels, we investigate the completeness (equivalent to the detection rate) of the Stixel World vs. the number of phantom Stixels. Among many findings, using this framework enables us to reduce the number of phantom Stixels by a factor of three compared to the base parametrization. This base parametrization has already been optimized by test driving vehicles for distances exceeding 10000 km.

show abstract