Vision-only 3D Tracking for Self-Driving Cars

Diaz-Ruiz, Carlos A.; Wang, Yan; Chao, Weilun; Weinberger, Kilian Q.; Campbell, Mark

doi:10.1109/coase.2019.8843260

Cited by 6 publications

(7 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Online 3D object tracking. We investigate this idea with a Kalman filter based tracker [61], [62], [63], which has shown promising results in benchmark tracking leader boards [12]. We opt to not use a learning-based tracker [64] since it would also require adaptation before it can be applied in the target domain.…”

Section: A Tracking For Improved Detectionmentioning

confidence: 99%

“…We opt to not use a learning-based tracker [64] since it would also require adaptation before it can be applied in the target domain. Specifically, we apply the tracker in [61]. The algorithm estimates the joint probability p(a k , x k |z k ) at time k, where x k is the set of tracked object states (e.g., cars speeds and locations), z k is the set of observed sensor measurements (here each measurement is a frame-wise detection), and a k is the assignment of measurements to tracks.…”

Section: A Tracking For Improved Detectionmentioning

confidence: 99%

“…Here we assume that the tracked vehicle is moving at a constant speed and heading in the global coordinate frame, with added noise to represent the uncertainty associated with vehicle maneuvers. This tracker has been shown to work well on tracking moving objects from a self-driving car [61], [65]. We initialize a new track when c min−hits measurements of the same tracked object are realized.…”

Section: A Tracking For Improved Detectionmentioning

confidence: 99%

“…The process noise matrix should represent the magnitudes of dynamical noise that the system might experience. In the model in [61], which is constant velocity and heading, there are seven noise parameters: e θ , e s , e vx , e vy , e ωz , e l , e w which correspond to the diagonal of the Q matrix. The variables e θ , e s , e l , e w are modeled as zero mean, mutually uncorrelated, Gaussian, and white process noise.…”

Section: A Hyper-parametersmentioning

confidence: 99%

“…5) The data association threshold (in BEV IoU) in GNN is set to be 0.3. 6) The fraction of distance between the vehicle center and back-axle is 1 4 l, as in [61]. • EKF parameters for extrapolation:…”

Section: A Hyper-parametersmentioning

confidence: 99%

See 4 more Smart Citations

Exploiting Playbacks in Unsupervised Domain Adaptation for 3D Object Detection

You

Diaz-Ruiz

Wang

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Self-driving cars must detect other vehicles and pedestrians in 3D to plan safe routes and avoid collisions. State-of-the-art 3D object detectors, based on deep learning, have shown promising accuracy but are prone to over-fit to domain idiosyncrasies, making them fail in new environmentsa serious problem if autonomous vehicles are meant to operate freely. In this paper, we propose a novel learning approach that drastically reduces this gap by fine-tuning the detector on pseudo-labels in the target domain, which our method generates while the vehicle is parked, based on replays of previously recorded driving sequences. In these replays, objects are tracked over time, and detections are interpolated and extrapolated-crucially, leveraging future information to catch hard cases. We show, on five autonomous driving datasets, that fine-tuning the object detector on these pseudo-labels substantially reduces the domain gap to new driving environments, yielding drastic improvements in accuracy and detection reliability.

show abstract