LaserFlow: Efficient and Probabilistic Object Detection and Motion Forecasting

Meyer, Gregory P.; Charland, Jake; Pandey, Shreyash; Laddha, Ankit P.; Gautam, Shivam Prakash; Vallespi-Gonzalez, Carlos; Wellington, Carl

doi:10.48550/arxiv.2003.05982

Cited by 5 publications

(11 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…MotionNet [32] used a spatial-temporal pyramid network to jointly perform detection and motion prediction for each BEV grid cell. LaserFlow [7] proposed an end-to-end model using multi-frame RV lidar inputs, unlike the other methods which use BEV representations, which can also perform prediction on multiple actor types. Compared to our method, most of the above end-to-end methods do not consider motion prediction on diverse road actor types, and none of them addresses the multimodal nature of possible future trajectories.…”

Section: Related Workmentioning

confidence: 99%

“…However, these approaches have generally focused only on vehicles and produce a single trajectory rather than full motion distributions. More recent work has shown the ability to learn a continuous distribution directly from sensor data for multiple classes [7], but the distributions are not multimodal. Prediction methods that operate on detections rather than the raw sensor data have shown improved performance by introducing multiclass predictions [8], estimates of uncertainty [9], [10], or incorporating multiple modes [11].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

MultiXNet: Multiclass Multistage Multimodal Motion Prediction

Djuric¹,

Cui²,

Su³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

One of the critical pieces of the self-driving puzzle is understanding the surroundings of the self-driving vehicle (SDV) and predicting how these surroundings will change in the near future. To address this task we propose MultiXNet, an end-to-end approach for detection and motion prediction based directly on lidar sensor data. This approach builds on prior work by handling multiple classes of traffic actors, adding a jointly trained second-stage trajectory refinement step, and producing a multimodal probability distribution over future actor motion that includes both multiple discrete traffic behaviors and calibrated continuous uncertainties. The method was evaluated on a large-scale, real-world data set collected by a fleet of SDVs in several cities, with the results indicating that it outperforms existing state-of-the-art approaches.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

MultiXNet: Multiclass Multistage Multimodal Motion Prediction

Djuric¹,

Cui²,

Su³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Fig. 2), optical flow [43], [60], depth estimation [50], [57], [61], [62], visual odometry [35], [63], [64], image annotation [41], [65]- [67], visual tracking [52], trajectory prediction [49], [68]- [70] and end-to-end perception [47], [71], [72]. Tab.…”

Section: Development Operationmentioning

confidence: 99%

A Review and Comparative Study on Probabilistic Object Detection in Autonomous Driving

Feng,

Harakeh,

Waslander

et al. 2020

Preprint

View full text Add to dashboard Cite

Capturing uncertainty in object detection is indispensable for safe autonomous driving. In recent years, deep learning has become the de-facto approach for object detection, and many probabilistic object detectors have been proposed. However, there is no summary on uncertainty estimation in deep object detection, and existing methods are not only built with different network architectures and uncertainty estimation methods, but also evaluated on different datasets with a wide range of evaluation metrics. As a result, a comparison among methods remains challenging, as does the selection of a model that best suits a particular application. This paper aims to alleviate this problem by providing a review and comparative study on existing probabilistic object detection methods for autonomous driving applications. First, we provide an overview of generic uncertainty estimation in deep learning, and then systematically survey existing methods and evaluation metrics for probabilistic object detection. Next, we present a strict comparative study for probabilistic object detection based on an image detector and three public autonomous driving datasets. Finally, we present a discussion of the remaining challenges and future works. Code has been made available at https://github.com/asharakeh/pod_ compare.git.

show abstract

“…The model assumes a univariate Laplace distribution for each of the (c x , c y , sin θ , cos θ ) components at each of the n + 1 timepoints, i.e, (n + 1) × 4 × 2 values are regressed for the means (µ's) and diversity parameters (b's). By contrast, in models with the polynomial representation, the regression values are the N µ + 1 coefficients in (4) and the N b + 1 coefficients in (5), for the four components individually. If multimodal prediction is modeled, such as for the vehicles in MultiXNet, independent polynomial representation is applied to each separate mode.…”

Section: Applying the Representation In Supervised Learningmentioning

confidence: 99%

“…In robotics in general and self-driving vehicle (SDV) applications in particular, anticipating the motion of other actors around the robot plays a critical role in planning safe paths to navigate the environment [1]. Recently, significant improvements have come from exploring the input representation of the sensor data [2], [3], [4], [5], [6] and the neural network structures [7], [8], [9]. Likewise, the output representation for trajectories has seen extensions to account for multimodality [10], [11], [12] and for modeling probability distributions over their future locations [9], [13].…”

Section: Introductionmentioning

confidence: 99%

Temporally-Continuous Probabilistic Prediction using Polynomial Trajectory Parameterization

Su¹,

Wang²,

Cui³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

A commonly-used representation for motion prediction of actors is a sequence of waypoints (comprising positions and orientations) for each actor at discrete future timepoints. While this approach is simple and flexible, it can exhibit unrealistic higher-order derivatives (such as acceleration) and approximation errors at intermediate time steps. To address this issue we propose a simple and general representation for temporally continuous probabilistic trajectory prediction that is based on polynomial trajectory parameterization. We evaluate the proposed representation on supervised trajectory prediction tasks using two large self-driving data sets. The results show realistic higher-order derivatives and better accuracy at interpolated time-points, as well as the benefits of the inferred noise distributions over the trajectories. Extensive experimental studies based on existing state-of-the-art models demonstrate the effectiveness of the proposed approach relative to other representations in predicting the future motions of vehicle, bicyclist, and pedestrian traffic actors.

show abstract

LaserFlow: Efficient and Probabilistic Object Detection and Motion Forecasting

Cited by 5 publications

References 28 publications

MultiXNet: Multiclass Multistage Multimodal Motion Prediction

MultiXNet: Multiclass Multistage Multimodal Motion Prediction

A Review and Comparative Study on Probabilistic Object Detection in Autonomous Driving

Temporally-Continuous Probabilistic Prediction using Polynomial Trajectory Parameterization

Contact Info

Product

Resources

About