车辆重识别
https://github.com/layumi/Vehicle_reID-Collection
数据集
vehicleX:https://github.com/yorkeyao/VehicleX(扩展数据集)
论文
论文主要从有代码的的和比较新的论文入手
2016
2017
会议
- Orientation Invariant Feature Embedding and Spatial Temporal Regularization for Vehicle Re-Identification (ICCV2017) [paper] [github]
2018
会议
期刊
workshop
2019
会议
workshop
- (Rank-1) Multi-camera vehicle tracking and re-identification based on visual and spatial-temporal features [[paper](http://openaccess.thecvf.com/content_CVPRW_2019/papers/AI City/Tan_Multi-camera_vehicle_tracking_and_re-identification_based_on_visual_and_spatial-temporal_CVPRW_2019_paper.pdf)] [github]
- (Rank-2) Multi-View Vehicle Re-Identification using Temporal Attention Model and Metadata Re-ranking [[paper](http://openaccess.thecvf.com/content_CVPRW_2019/papers/AI City/Huang_Multi-View_Vehicle_Re-Identification_using_Temporal_Attention_Model_and_Metadata_Re-ranking_CVPRW_2019_paper.pdf)] [github]
- (Rank-3) Vehicle Re-identification with Location and Time Stamps [[paper](http://openaccess.thecvf.com/content_CVPRW_2019/papers/AI City/Lv_Vehicle_Re-Identification_with_Location_and_Time_Stamps_CVPRW_2019_paper.pdf)] [github]
- (Rank-5) Multi-Camera Vehicle Tracking with Powerful Visual Features and Spatial-Temporal Cue [[paper](http://openaccess.thecvf.com/content_CVPRW_2019/papers/AI City/He_Multi-Camera_Vehicle_Tracking_with_Powerful_Visual_Features_and_Spatial-Temporal_Cue_CVPRW_2019_paper.pdf)] [github]
- (Rank-25) Vehicle Re-identification with Learned Representation and Spatial Verification and Abnormality Detection with Multi-Adaptive Vehicle Detectors for Traffic Video Analysis [[paper](http://openaccess.thecvf.com/content_CVPRW_2019/papers/AI City/Nguyen_Vehicle_Re-identification_with_Learned_Representation_and_Spatial_Verification_and_Abnormality_CVPRW_2019_paper.pdf)] [github]
- (Rank-50) AI City Challenge 2019 – City-Scale Video Analytics for Smart Transportation [[paper](http://openaccess.thecvf.com/content_CVPRW_2019/papers/AI City/Chang_AI_City_Challenge_2019_–_City-Scale_Video_Analytics_for_Smart_CVPRW_2019_paper.pdf)] [github]
2020
期刊
- VehicleNet: Learning Robust Visual Representation for Vehicle Re-identification (TMM2020) [paper]
workshop
- (Rank-1) Going Beyond Real Data: A Robust Visual Representation for Vehicle Re-Identification [paper] [github]
- (Rank-2) VOC-ReID: Vehicle Re-Identification Based on Vehicle-Orientation-Camera [paper] [github]
- (Rank-3) Multi-Domain Learning and Identity Mining for Vehicle Re-Identification [paper] [github]
- (Rank-4) Large Scale Vehicle Re-Identification by Knowledge Transfer From Simulated Data and Temporal Attention [paper] [github]
- (Rank-26) iTASK - Intelligent Traffic Analysis Software Kit [paper] [github]
- (General) Vehicle Re-Identification Based on Complementary Features [paper] [github]
- (General) Attribute-Guided Feature Extraction and Augmentation Robust Learning for Vehicle Re-Identification [paper] [github]
AI城市挑战赛
官网地址: https://www.aicitychallenge.org/2020-challenge-tracks/
主要比赛项目:
Detailed participant instructions can be accessed here.
Participants can compete in one or more of the following four challenges:
Challenge Track 1: Multi-Class Multi-Movement Vehicle Counting
Participating teams will count four-wheel vehicles and freight trucks that follow pre-defined movements from multiple camera scenes. For example, teams will perform vehicle counting separately for left-turning, right-turning and through traffic at a given intersection approach. This helps traffic engineers understand the traffic demand and freight ratio on individual corridors, which can be used to design better intersection signal timing plans and apply other traffic congestion mitigation strategies when necessary. To maximize the practical value of the outcome from this track, both the vehicle counting effectiveness and the program execution efficiency will contribute to the final score for each participating team. The team with the highest score will be declared the winner of this track.
Challenge Track 2: City-Scale Multi-Camera Vehicle Re-Identification
Participating teams will perform vehicle re-identification based on vehicle crops from multiple cameras placed at multiple intersections. This helps traffic engineers understand journey times along entire corridors. In this year’s challenge, the training set will be composed of both real-world data and synthetic data. The usage of synthetic data is encouraged as it can be simulated under various environments and can produce large training data sets. The team with the highest accuracy in detecting vehicles that appear in multiple cameras will be declared the winner of this track. In the event, that multiple teams perform equally well in this track, the algorithm needing the least amount of manual supervision will be chosen as the winner.
Challenge Track 3: City-Scale Multi-Camera Vehicle Tracking
Participating teams will track vehicles across multiple cameras both at a single intersection and across multiple intersections spread out across a city. This helps traffic engineers understand journey times along entire corridors. The team with the highest accuracy in detecting vehicles that appear in multiple cameras will be declared the winner of this track.In the event that multiple teams perform equally well in this track, the algorithm needing the least amount of manual supervision will be chosen as the winner.
Challenge Track 4: Traffic Anomaly Detection
Participating teams will submit at most 100 anomalies detected, including wrong turns, wrong driving direction, lane change errors, and all other anomalies, based on video feeds available from multiple cameras at intersections and along highways. The team with the highest average precision in anomaly detection in the submitted anomalies will be announced the winner of this track.
------------------翻译如下---------------------
可以在此处访问详细的参与者说明。
参与者可以参加以下四个挑战中的一项或多项:
挑战赛1:多级多运动车辆计数
参加团队将计算遵循多摄像机场景中预定义动作的四轮车辆和货车。例如,在给定的交叉路口,车队将分别为左转,右转和通过交通进行车辆计数。这有助于交通工程师了解各个走廊上的交通需求和货运比率,可用于设计更好的十字路口信号计时计划,并在必要时应用其他缓解交通拥堵的策略。为了最大程度地提高此赛道成果的实用价值,车辆计数效率和程序执行效率都将为每个参赛团队的最终得分做出贡献。得分最高的团队将被宣布为该赛段的冠军。
挑战之路2:城市规模的多摄像机车辆重新识别
参加团队将根据来自多个交叉路口的多个摄像机的车辆收成进行车辆重新识别。这有助于交通工程师了解整个走廊的旅行时间。在今年的挑战中,训练集将由现实数据和综合数据组成。鼓励使用合成数据,因为它可以在各种环境下进行模拟,并且可以生成大量的训练数据集。检测多辆摄像机中出现的车辆的准确性最高的团队将被宣布为该赛道的获胜者。如果有多个团队在此赛道上表现出色,则需要最少人工监督的算法将被选为获胜者。
挑战赛3:城市规模多摄像机车辆跟踪
参与团队将在单个路口以及分布在城市中的多个路口处,通过多个摄像机跟踪车辆。这有助于交通工程师了解整个走廊的旅行时间。检测多辆摄像机中出现的车辆的准确性最高的团队将被宣布为该赛道的获胜者。如果多个团队在该赛道中的表现均相同,则将选择需要最少人工监督的算法作为优胜者。
挑战四:交通异常检测
参与团队将根据交叉路口和高速公路上多个摄像机提供的视频源,提交最多100个检测到的异常,包括错误的转弯,错误的驾驶方向,换道错误以及所有其他异常。在提交的异常中异常检测中平均精度最高的团队将宣布为该赛道的冠军。
2020年时间表(相当于是1月开始,一直持续到6月份)
Important Dates 2020
Challenge kick off: Tuesday, Jan 7Data sets shared with participants: Friday, Jan 10Evaluation server open to submissions: Saturday, March 7Challenge track submissions due: Thursday, April 9 (11:59 PM, Pacific Time)
Evaluation submission is closed and rankings are finalized.Workshop papers due: Monday, April 13 (09:00 AM, Pacific Time)
Since our review is not double-blind, papers should be submitted in final/camera-ready form.Final decisions to authors: Saturday, April 18
All authors are notified in CMT. There are about 24 hours to prepare for the final version of accepted papers.Final papers due: Sunday, April 19 (11:59 PM, Pacific Time)
All camera ready paper should be uploaded to CMT to be published by CVPR 2020. The accepted workshop papers will be accessible online at IEEE Xplore Digital Library and CVF Open Access.Open source on GitHub (training code + testing code + additional annotation) due: Sunday, May 10 (11:59 PM, Pacific Time)
All the competitors/candidates for awards MUST release their code for validation before decision of awardees. The performance on the leaderboard has to be reproducible without the use of external data.- Presentation of papers and announcement of awards: Monday, June 15
数据集
Specifically, the datasets provided for the challenge this year are CityFlow [37, 24] (Track 2 ReID and Track 3 MTMC), VehicleX [40, 36] (Track 2), Iowa DOT [23] dataset (Track 4 anomaly event detection and Track 1 vehicle counting).