|
|
Recently, over-height automobile strike regularly happens, causing nice economic value and critical safety problems. Hence, an alert system which can accurately uncover any possible top limiting gadgets prematurely is necessary to be employed in modern large or medium sized cars, akin to touring automobiles. Detecting and estimating the height limiting gadgets act as the important thing point of a profitable top restrict alert system. Though there are some works research height limit estimation, existing methods are either too computational expensive or not accurate sufficient. On this paper, we propose a novel stereo-based mostly pipeline named SHLE for top limit estimation. Our SHLE pipeline consists of two stages. In stage 1, a novel devices detection and tracking scheme is introduced, which accurately locate the height restrict units in the left or proper image. Then, in stage 2, the depth is temporally measured, extracted and filtered to calculate the height limit gadget. To benchmark the height limit estimation job, we construct a large-scale dataset named "Disparity Height", the place stereo pictures, pre-computed disparities and ground-fact height restrict annotations are provided.
my blog; buy itagpro |
|
發表於