RoadTrack: Realtime Tracking of Road Agents in Dense and Heterogeneous Environments

Chandra, Rohan; Bhattacharya, Uttaran; Randhavane, Tanmay; Bera, Aniket; Manocha, Dinesh

doi:10.1109/icra40945.2020.9196612

Cited by 12 publications

(10 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Ids can be explained as the miss-switch of the instances of the objects at different time steps. If the object detector can detect the objects accurately for a given domain, the performance of tracking-by-detection can also be sufficient [5,15]; however, the object detector cannot always be accurate for partially observed objects.…”

Section: Related Workmentioning

confidence: 99%

“…Bayesian filters have a high capacity of predicting the occupancy and motion states of the grids, which are not directly observable by the sensors [22]. In MOT, when Bayesian filters are used, they are generally applied on the object detections to predict the next location of the detected object ( [28,6,5]). We are proposing a new approach for MOT by integrating the Bayesian filter on raw sensor data with different off-the-shelf precise object detectors ( [13,26]).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

GridTrack: Detection and Tracking of Multiple Objects in Dynamic Occupancy Grids

Erkent

González

Paigwar

et al. 2021

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Multiple Object Tracking is an important task for autonomous vehicles. However, it gets difficult to track objects when it is hard to detect them due to occlusion or distance to the sensors. We propose a method, "GridTrack", to overcome this difficulty. We fuse a dynamic occupancy grid map (DOGMa) with an object detector. DOGMa is obtained by applying a Bayesian filter on raw sensor data. This improves the tracking of the partially observed / unobserved objects with the help of the Bayesian filter on raw data, which has a powerful prediction capability. We develop a network to track the objects on the grid and fuse information from previous detections in this network. The experiments show that the multi-object tracking accuracy is high with the usage of the proposed method.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

GridTrack: Detection and Tracking of Multiple Objects in Dynamic Occupancy Grids

Erkent

González

Paigwar

et al. 2021

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…Deep learning-based methods [30] train a recurrent neural network for trajectory prediction [7], [11], [12] and tracking [6], [10] are also susceptible to complex environments or different behavior of drivers. In contrast to all of the above, GAMEPLAN do not require an objective function; instead, successful application of this approach requires minimizing a loss function that depends on the distribution of the given data.…”

Section: Prior Workmentioning

confidence: 99%

GamePlan: Game-Theoretic Multi-Agent Planning with Human Drivers at Intersections, Roundabouts, and Merging

Chandra¹,

Manocha²

2021

Preprint

Self Cite

View full text Add to dashboard Cite

We present a new method for multi-agent planning involving human drivers and autonomous vehicles (AVs) in unsignaled intersections, roundabouts, and during merging. In multi-agent planning, the main challenge is to predict the actions of other agents, especially human drivers, as their intentions are hidden from other agents. Our algorithm uses game theory to develop a new auction, called GAMEPLAN, that directly determines the optimal action for each agent based on their driving style (which is observable via commonly available sensors like lidars and cameras). GAMEPLAN assigns a higher priority to more aggressive or impatient drivers and a lower priority to more conservative or patient drivers; we theoretically prove that such an approach, although counter-intuitive, is gametheoretically optimal. Our approach successfully prevents collisions and deadlocks. We compare our approach with prior state-of-the-art auction techniques including economic auctions, time-based auctions (first-in first-out), and random bidding and show that each of these methods result in collisions among agents when taking into account driver behavior. We additionally compare with methods based on deep reinforcement learning, deep learning, and game theory and present our benefits over these approaches. Finally, we show that our approach can be implemented in the real-world with human drivers.

show abstract

“…Datasets such as the Argoverse [13], Lyft Level 5 [22], Waymo Open Dataset [18], ApolloScape [25], nuScenes dataset [2] have been used for trajectory forecasting [7], [11], [12], [20], [1], [34], [4], [29] and tracking [6], [10]. We summarize various characteristics of our dataset in terms of scene: traffic density, road type, lighting conditions, agents (we indicate the total count of each agent across 1250 videos), and behaviors, along with their size distribution (in GB).…”

Section: A Tracking and Trajectory Predictionmentioning

confidence: 99%

METEOR:A Dense, Heterogeneous, and Unstructured Traffic Dataset With Rare Behaviors

Chandra¹,

Wang²,

Mahajan³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

We present a new and complex traffic dataset, METEOR, which captures traffic patterns in unstructured scenarios in India. METEOR consists of more than 1000 oneminute video clips, over 2 million annotated frames with egovehicle trajectories, and more than 13 million bounding boxes for surrounding vehicles or traffic agents. METEOR is a unique dataset in terms of capturing the heterogeneity of microscopic and macroscopic traffic characteristics. Furthermore, we provide annotations for rare and interesting driving behaviors such as cut-ins, yielding, overtaking, overspeeding, zigzagging, sudden lane changing, running traffic signals, driving in the wrong lanes, taking wrong turns, lack of right-of-way rules at intersections, etc. We also present diverse traffic scenarios corresponding to rainy weather, nighttime driving, driving in rural areas with unmarked roads, and high-density traffic scenarios. We use our novel dataset to evaluate the performance of object detection and behavior prediction algorithms. We show that state-of-theart object detectors fail in these challenging conditions and also propose a new benchmark test: action-behavior prediction with a baseline mAP score of 70.74. I. INTRODUCTIONRecent research in learning-based techniques for robotics, computer vision, and autonomous driving has been driven by the availability of datasets and benchmarks. Several traffic datasets have been collected from different parts of the world to stimulate research in autonomous driving, driver assistants, and intelligent traffic systems. These datasets correspond to highway traffic or urban traffic and are widely used in the development and evaluation of new methods for perception [6], [10], prediction [7], [11], [12], [20], [1], behavior analysis [5], [8], [9], and navigation [28], [26].Many initial autonomous driving datasets were motivated by computer vision or perception tasks such as object recognition, semantic segmentation or 3D scene understanding. Recently, many other datasets have been released that consist of pointcloud representations of objects captured using LiDAR, pose information, 3D track information, stereo imagery or detailed map information for applications related to 3D object recognition and motion forecasting. Many large-scale motion forecasting datasets such as Argoverse [13], and Waymo Open Motion Dataset [18], among others, have been used extensively by researchers and engineers to develop robust prediction

show abstract

RoadTrack: Realtime Tracking of Road Agents in Dense and Heterogeneous Environments

Cited by 12 publications

References 60 publications

GridTrack: Detection and Tracking of Multiple Objects in Dynamic Occupancy Grids

GridTrack: Detection and Tracking of Multiple Objects in Dynamic Occupancy Grids

GamePlan: Game-Theoretic Multi-Agent Planning with Human Drivers at Intersections, Roundabouts, and Merging

METEOR:A Dense, Heterogeneous, and Unstructured Traffic Dataset With Rare Behaviors

Contact Info

Product

Resources

About