A Hierarchical Network for Diverse Trajectory Proposals

Sriram, N N; Kumar, Gourav; Singh, Abhay Pratap; Karthik, M. Siva; Saurav, Saket; Bhowrnick, Brojeshwar; Krishna, K. Madhava

doi:10.1109/ivs.2019.8813986

Cited by 4 publications

(2 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Multi-Choice Learning: Multi-modal predictions have been realized in different domains through Multi-choice learning (MCL) [19,13,30] objectives in the past. Several works have shown use cases of MCL to provide diverse hypotheses in classification [30,39], segmentation [30,39], captioning [30], pose estimation [39], image synthesis [11] and trajectory proposals [45]. Convergence issue related to WTA objectives have been shown in [39,34].…”

Section: Related Workmentioning

confidence: 99%

Divide-and-Conquer for Lane-Aware Diverse Trajectory Prediction

Narayanan¹,

Moslemi²,

Pittaluga³

et al. 2021

Preprint

View full text Add to dashboard Cite

Trajectory prediction is a safety-critical tool for autonomous vehicles to plan and execute actions. Our work addresses two key challenges in trajectory prediction, learning multimodal outputs, and better predictions by imposing constraints using driving knowledge. Recent methods have achieved strong performances using Multi-Choice Learning objectives like winner-takes-all (WTA) or best-of-many. But the impact of those methods in learning diverse hypotheses is under-studied as such objectives highly depend on their initialization for diversity. As our first contribution, we propose a novel Divide-And-Conquer (DAC) approach that acts as a better initialization technique to WTA objective, resulting in diverse outputs without any spurious modes. Our second contribution is a novel trajectory prediction framework called ALAN that uses existing lane centerlines as anchors to provide trajectories constrained to the input lanes. Our framework provides multi-agent trajectory outputs in a forward pass by capturing interactions through hypercolumn descriptors and incorporating scene information in the form of rasterized images and per-agent lane anchors. Experiments on synthetic and real data show that the proposed DAC captures the data distribution better compare to other WTA family of objectives. Further, we show that our ALAN approach provides on par or better performance with SOTA methods evaluated on Nuscenes urban driving benchmark.

show abstract

Section: Related Workmentioning

confidence: 99%

Divide-and-Conquer for Lane-Aware Diverse Trajectory Prediction

Narayanan¹,

Moslemi²,

Pittaluga³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…While more advanced methods [20,16] show outputs that are tightly coupled with the ground truth. We do not intend to capture the data distribution in such a fashion but are more focused towards producing predictions in possible dominant choices of motion [35]. This also motivates us to use [14] to showcase the ability of simulation strategy in producing more diverse outputs.…”

Section: Related Workmentioning

confidence: 99%

SMART: Simultaneous Multi-Agent Recurrent Trajectory Prediction

Liu

Pittaluga

et al. 2020

Computer Vision – ECCV 2020

Self Cite

View full text Add to dashboard Cite

We propose advances that address two key challenges in future trajectory prediction: (i) multimodality in both training data and predictions and (ii) constant time inference regardless of number of agents. Existing trajectory predictions are fundamentally limited by lack of diversity in training data, which is difficult to acquire with sufficient coverage of possible modes. Our first contribution is an automatic method to simulate diverse trajectories in the top-view. It uses pre-existing datasets and maps as initialization, mines existing trajectories to represent realistic driving behaviors and uses a multi-agent vehicle dynamics simulator to generate diverse new trajectories that cover various modes and are consistent with scene layout constraints. Our second contribution is a novel method that generates diverse predictions while accounting for scene semantics and multi-agent interactions, with constant-time inference independent of the number of agents. We propose a convLSTM with novel state pooling operations and losses to predict scene-consistent states of multiple agents in a single forward pass, along with a CVAE for diversity. We validate our proposed multi-agent trajectory prediction approach by training and testing on the proposed simulated dataset and existing real datasets of traffic scenes. In both cases, our approach outperforms SOTA methods by a large margin, highlighting the benefits of both our diverse dataset simulation and constant-time diverse trajectory prediction methods.

show abstract