泡泡一分钟:BLVD: Building A Large-scale 5D Semantics Benchmark for Autonomous Driving
BLVD: Building A Large-scale 5D Semantics Benchmark for Autonomous Driving
BLVD:構(gòu)建自主駕駛的大規(guī)模5D語(yǔ)義基準(zhǔn)
Jianru Xue, Jianwu Fang, Tao Li, Bohua Zhang, Pu Zhang, Zhen Ye and Jian Dou
Abstract—In autonomous driving community, numerous benchmarks have been established to assist the tasks of 3D/2D object detection, stereo vision, semantic/instance segmentation. However, the more meaningful dynamic evolution of the surrounding objects of ego-vehicle is rarely exploited, and lacks a large-scale dataset platform. To address this, we introduce BLVD, a large-scale 5D semantics benchmark which does not concentrate on the static detection or semantic/instance segmentation tasks tackled adequately before. Instead, BLVD aims to provide a platform for the tasks of dynamic 4D (3D+temporal) tracking, 5D (4D+interactive) interactive event recognition and intention prediction.This benchmark will boost the deeper understanding of traf?c scenes than ever before. We totally yield 249,129 3D annotations, 4,902 independent individuals for tracking with the length of overall 214,922 points, 6,004 valid fragments for 5D interactive event recognition, and 4,900 individuals for 5D intention prediction. These tasks are contained in four kinds of scenarios depending on the object density (low and high) and light conditions (daytime and nighttime). The benchmark can be downloaded from our project site https://github.com/VCCIV/BLVD/.
在自動(dòng)駕駛社區(qū)中,已經(jīng)建立了許多基準(zhǔn)來(lái)輔助3D / 2D物體檢測(cè),立體視覺,語(yǔ)義/實(shí)例分割的任務(wù)。然而,自我車輛周圍物體的更有意義的動(dòng)態(tài)演化很少被利用,并且缺乏大規(guī)模的數(shù)據(jù)集平臺(tái)。為了解決這個(gè)問(wèn)題,我們引入了BLVD,這是一個(gè)大規(guī)模的5D語(yǔ)義基準(zhǔn)測(cè)試,它不專注于之前充分處理的靜態(tài)檢測(cè)或語(yǔ)義/實(shí)例分割任務(wù)。相反,BLVD旨在為動(dòng)態(tài)4D(3D +時(shí)間)跟蹤,5D(4D +交互式)交互式事件識(shí)別和意圖預(yù)測(cè)的任務(wù)提供平臺(tái)。該基準(zhǔn)將比以往更加深入地了解交通場(chǎng)景。 我們完全產(chǎn)生249,129個(gè)3D注釋,4,902個(gè)獨(dú)立個(gè)體用于跟蹤,總長(zhǎng)度為214,922個(gè)點(diǎn),6,004個(gè)有效片段用于5D交互事件識(shí)別,4,900個(gè)用于5D意圖預(yù)測(cè)。這些任務(wù)包含在四種場(chǎng)景中,具體取決于對(duì)象密度(低和高)和光照條件(白天和夜晚)。 基準(zhǔn)測(cè)試可以從我們的項(xiàng)目站點(diǎn)https://github.com/VCCIV/BLVD/下載。
在本文中,我們?yōu)樽詣?dòng)駕駛構(gòu)建了一個(gè)大規(guī)模的5D語(yǔ)義基準(zhǔn),該基準(zhǔn)在各種有趣的場(chǎng)景下被捕獲,并且經(jīng)過(guò)有效和準(zhǔn)確的校準(zhǔn),同步和整流。與以前的靜態(tài)檢測(cè)/分割任務(wù)不同,我們專注于對(duì)交通場(chǎng)景的更深入理解。具體而言,4D跟蹤,5D交互事件識(shí)別和5D意圖預(yù)測(cè)的任務(wù)在該基準(zhǔn)測(cè)試中啟動(dòng)。通過(guò)仔細(xì)的注釋,基準(zhǔn)產(chǎn)生了249,129個(gè)3D注釋,4,902個(gè)獨(dú)立實(shí)例用于跟蹤,總長(zhǎng)度為214,922個(gè)點(diǎn),6,004個(gè)用于5D交互式事件識(shí)別的3D注釋,以及4,900個(gè)用于5D意圖預(yù)測(cè)的個(gè)體。這些注釋是在不同的光照條件下(白天和夜晚),不同密度的參與者(低密度和高密度)和不同的駕駛場(chǎng)景(高速公路和城市)收集的。
?
轉(zhuǎn)載于:https://www.cnblogs.com/feifanrensheng/p/11368354.html
總結(jié)
以上是生活随笔為你收集整理的泡泡一分钟:BLVD: Building A Large-scale 5D Semantics Benchmark for Autonomous Driving的全部?jī)?nèi)容,希望文章能夠幫你解決所遇到的問(wèn)題。
- 上一篇: 【宋红康 MySQL数据库 】【高级篇】
- 下一篇: 微观:心流,宏观:ikigai