Haar Wavelet Based Block Autoregressive Flows for Trajectories

@article{Bhattacharyya2020HaarWB,
  title={Haar Wavelet Based Block Autoregressive Flows for Trajectories},
  author={Apratim Bhattacharyya and Christoph Nikolas Straehle and Mario Fritz and Bernt Schiele},
  journal={Pattern Recognition},
  year={2020},
  volume={12544},
  pages={275 - 288}
}
Prediction of trajectories such as that of pedestrians is crucial to the performance of autonomous agents. While previous works have leveraged conditional generative models like GANs and VAEs for learning the likely future trajectories, accurately modeling the dependency structure of these multimodal distributions, particularly over long time horizons remains challenging. Normalizing flow based generative models can model complex distributions admitting exact inference. These include variants… Expand

Figures and Tables from this paper

Trajectory Prediction in Autonomous Driving With a Lane Heading Auxiliary Loss
TLDR
A new metric is proposed which addresses failure cases of the off-road rate metric by penalizing trajectories that oppose the ascribed heading of a driving lane, and this metric is shown to be differentiable and therefore suitable as an auxiliary loss function. Expand
Euro-PVI: Pedestrian Vehicle Interactions in Dense Urban Centers
TLDR
This work proposes Euro-PVI, a dataset of pedestrian and bicyclist trajectories that caters more diverse and complex interactions in dense urban scenarios compared to the existing datasets and develops a joint inference model that learns an expressive multi-modal shared latent space across agents in the urban scene. Expand
FloMo: Tractable Motion Prediction with Normalizing Flows
TLDR
This work model motion prediction directly as a density estimation problem with a normalizing flow between a noise sample and the future motion distribution and achieves state-of-the-art performance on three popular prediction datasets, with a significant gap to most competing models. Expand
Learning Game-Theoretic Models of Multiagent Trajectories Using Implicit Layers
TLDR
An end-to-end trainable architecture is proposed that hybridizes neural nets with game-theoretic reasoning, has interpretable intermediate representations, and transfers to downstream decision making, and learns an equilibrium refinement concept. Expand
Trajectory Forecasts in Unknown Environments Conditioned on Grid-Based Plans
TLDR
An attention based trajectory generator that generates continuous valued future trajectories conditioned on state sequences sampled from the MaxEnt policy is proposed that generates trajectories that are diverse, representing the multi-modal predictive distribution, and precise, conforming to the underlying scene structure over long prediction horizons. Expand

References

SHOWING 1-10 OF 44 REFERENCES
Normalizing Flows With Multi-Scale Autoregressive Priors
TLDR
The mAR prior for models with split coupling flow layers (mAR-SCF) can better capture dependencies in complex multimodal data and achieves state-of-the-art density estimation results on MNIST, CIFAR-10, and ImageNet. Expand
Latent Normalizing Flows for Discrete Sequences
TLDR
A VAE-based generative model is proposed which jointly learns a normalizing flow-based distribution in the latent space and a stochastic mapping to an observed discrete space in this setting, finding that it is crucial for the flow- based distribution to be highly multimodal. Expand
Conditional Flow Variational Autoencoders for Structured Sequence Prediction
TLDR
This work introduces Conditional Flow Variational Autoencoders (CF-VAE) and proposes two novel regularization schemes which stabilizes training and deals with posterior collapse for stable training and better fit to the target data distribution. Expand
Multi-Agent Tensor Fusion for Contextual Trajectory Prediction
  • Tianyang Zhao, Yifei Xu, +5 authors Y. Wu
  • Computer Science
  • 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2019
TLDR
This work encodes multiple agents' past trajectories and the scene context into a Multi-Agent Tensor, then applies convolutional fusion to capture multiagent interactions while retaining the spatial structure of agents and thescene context. Expand
Diverse Trajectory Forecasting with Determinantal Point Processes
TLDR
This work proposes to learn a diversity sampling function (DSF) that generates a diverse and likely set of future trajectories and demonstrates the diversity of the trajectories produced by the approach on both low-dimensional 2D trajectory data and high-dimensional human motion data. Expand
Convolutional Social Pooling for Vehicle Trajectory Prediction
  • Nachiket Deo, M. Trivedi
  • Computer Science
  • 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
  • 2018
TLDR
This paper proposes an LSTM encoder-decoder model that uses convolutional social pooling as an improvement to social Pooling layers for robustly learning interdependencies in vehicle motion and outputs a multi-modal predictive distribution over future trajectories based on maneuver classes. Expand
r2p2: A ReparameteRized Pushforward Policy for Diverse, Precise Generative Path Forecasting
TLDR
A method to forecast a vehicle’s ego-motion as a distribution over spatiotemporal paths, conditioned on features embedded in an overhead map, and obtains expressions for the cross-entropy metrics that can be efficiently evaluated and differentiated, enabling stochastic-gradient optimization. Expand
Social LSTM: Human Trajectory Prediction in Crowded Spaces
TLDR
This work proposes an LSTM model which can learn general human movement and predict their future trajectories and outperforms state-of-the-art methods on some of these datasets. Expand
CAR-Net: Clairvoyant Attentive Recurrent Network
TLDR
A Clairvoyant Attentive Recurrent Network (CAR-Net) that learns where to look in a large image of the scene when solving the path prediction task, and shows CAR-Net’s ability to generalize to unseen scenes. Expand
Flow++: Improving Flow-Based Generative Models with Variational Dequantization and Architecture Design
TLDR
Flow++ is proposed, a new flow-based model that is now the state-of-the-art non-autoregressive model for unconditional density estimation on standard image benchmarks, and has begun to close the significant performance gap that has so far existed between autoregressive models and flow- based models. Expand
...
1
2
3
4
5
...