The KITTI Vision Benchmark Suite之Stereo Evaluation 2012
The stereo / flow benchmark consists of 194 training image pairs and 195 test image pairs, saved in loss less png format. Our evaluation server computes the average number of bad pixels for all non-occluded or occluded (=all groundtruth) pixels. We require
that all methods use the same parameter set for all test pairs. Our development kit provides details about the data format as well as MATLAB / C++ utility functions for reading and writing disparity maps and flow fields.
Our evaluation table ranks all methods according to the number of non-occluded erroneous pixels at the specified disparity / end-point error threshold. All methods providing less than 100 % density have beeninterpolated using simple background interpolation as explained in the corresponding header file in the development kit. For each method we show:
- Out-Noc: Percentage of erroneous pixels innon-occluded areas
- Out-All: Percentage of erroneous pixels in total
- Avg-Noc: Average disparity / end-point error in non-occluded areas
- Avg-All: Average disparity / end-point error in total
- Density: Percentage of pixels for which ground truth has been provided by the method
Note: On 04.11.2013 we have improved theground truth disparity maps andflow fields leading to slightly improvements for all methods. Please download the stereo/flow dataset with theimproved ground truth for training again, if you have downloaded the dataset prior to 04.11.2013. Please consider reporting these new number for all future submissions. Links to last leaderboards排行榜 before the updates: stereo and flow!
- Flow: Method uses optical flow (2 temporally adjacent images)
- Multiview: Method uses more than 2 temporally adjacent images
- Motion stereo: Method usesepipolar geometry 核面幾何學for computing optical flow
- Additional training data: Use of additional data sources for training (see details)