ScePT: Scene-consistent, Policy-based Trajectory Predictions for Planning

Chen, Yuxiao; Ivanovic, Boris; Pavone, Marco

doi:10.1109/cvpr52688.2022.01659

Cited by 33 publications

(37 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The representative trajectory prediction frameworks applying these designs are, e.g., Social-GAN [34], DESIRE [35], and Precog [36], respectively. Due to the relatively easy training process and good performance, many recent works [37], [13], [38], [39] extend the CVAE-based design for multimodal trajectory prediction. However, these sampling-based approaches do not provide a straightforward mechanism to estimate the likelihood of each prediction in the random sampling process.…”

Section: Related Workmentioning

confidence: 99%

“…To make a fair comparison, we only benchmark our model GATraj with multi-path trajectory prediction models. Namely, they are convolutional based multi-path trajectory prediction models CoverNet [5] and MTP [44]; GANbased model Social-GAN [34]; CVAE-based models DLow-AF [45] and ScePT [39]; Normalizing flow-based model LDS-AF [46]; GCN-based models STAR [47]; GMM-based models MultiPath [48] and Trajectron++ [40]; attentionbased models SoPhie [10], AgentFormer [13], MID [26]; clustering-based model PCCSNet [49]; belief energy-based model LB-EBM [50]. It should be noted that some of the baseline models [10], [44], [48] also include scene information.…”

Section: B Evaluation Metrics and Baselinesmentioning

confidence: 99%

“…1, recent deep learning models trained on large-scale real-world datasets have significantly improved trajectory prediction accuracy on multiple benchmarks measured by distance errors. These models address [40] PCCSNet [49] STAR [47] AgentFormer [13] ScePT [39] Fig. 1: The inference speed and the prediction performance of the models on the Zara2 dataset [1] computed from a single Tesla V100 GPU.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

GATraj: A Graph- and Attention-based Multi-Agent Trajectory Prediction Model

Cheng¹,

Liu²,

Chen³

et al. 2022

Preprint

View full text Add to dashboard Cite

Trajectory prediction has been a long-standing problem in intelligent systems such as autonomous driving and robot navigation. Recent state-of-the-art models trained on large-scale benchmarks have been pushing the limit of performance rapidly, mainly focusing on improving prediction accuracy. However, those models put less emphasis on efficiency, which is critical for real-time applications. This paper proposes an attention-based graph model named GATraj with a much higher prediction speed. Spatial-temporal dynamics of agents, e.g., pedestrians or vehicles, are modeled by attention mechanisms. Interactions among agents are modeled by a graph convolutional network. We also implement a Laplacian mixture decoder to mitigate mode collapse and generate diverse multimodal predictions for each agent. Our model achieves performance on par with the state-of-the-art models at a much higher prediction speed tested on multiple open datasets.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: B Evaluation Metrics and Baselinesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

GATraj: A Graph- and Attention-based Multi-Agent Trajectory Prediction Model

Cheng¹,

Liu²,

Chen³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…We observe a proliferation of algorithms [32,31,23,42,27] that try to optimize the path prediction for a single context and perspective. Most existing approaches focus on an isolated domain and fail to evaluate the generality of their solution over different contexts and domains [4,25,31]. It is rare to see approaches that predict for both vehicles and pedestrians [30,4], and it is only recently that select pedestrian focused approaches started using both bird's-eye and high-angle views [11,22,21,20,17].…”

Section: Introductionmentioning

confidence: 99%

“…Most existing approaches focus on an isolated domain and fail to evaluate the generality of their solution over different contexts and domains [4,25,31]. It is rare to see approaches that predict for both vehicles and pedestrians [30,4], and it is only recently that select pedestrian focused approaches started using both bird's-eye and high-angle views [11,22,21,20,17]. When it comes to real-world deployment and applications, a general domainagnostic approach that can accurately predict both pedestrians and vehicles across different views and perspectives is highly desirable.…”

Section: Introductionmentioning

confidence: 99%

Pishgu: Universal Path Prediction Architecture through Graph Isomorphism and Attentive Convolution

Noghre¹,

Vinit²,

Pazho³

et al. 2022

Preprint

View full text Add to dashboard Cite

Path prediction is an essential task for several real-world real-time applications, from autonomous driving and video surveillance to environmental monitoring. Most existing approaches are computation-intensive and only target a narrow domain (e.g., a specific point of view for a particular subject). However, many real-time applications demand a universal path predictor that can work across different subjects (vehicles, pedestrians), perspectives (bird's-eye, highangle), and scenes (sidewalk, highway). This article proposes Pishgu, a universal graph isomorphism approach for attentive path prediction that accounts for environmental challenges. Pishgu captures the inter-dependencies within the subjects in each frame by taking advantage of Graph Isomorphism Networks. In addition, an attention module is adopted to represent the intrinsic relations of the subjects of interest with their surroundings. We evaluate the adaptability of our approach to multiple publicly available vehicle (bird's-eye view) and pedestrian (bird's-eye and high-angle view) path prediction datasets. Pishgu's universal solution outperforms existing domain-focused methods by producing state-of-the-art results for vehicle bird's-eye view by 42% and 61% and pedestrian high-angle views by 23% and 22% in terms of ADE and FDE, respectively. Moreover, we analyze the domain-specific details for various datasets to understand their effect on path prediction and model interpretation. Although our model is a single solution for path prediction problems and defines a new standard in multiple domains, it still has a comparable complexity to state-ofthe-art models, which makes it suitable for real-world application. We also report the latency and throughput for all three domains on multiple embedded processors.

show abstract

View Vertically: A Hierarchical Network for Trajectory Prediction via Fourier Spectrums

Wong

Xia

Hong

et al. 2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

With the fast development of AI-related techniques, the applications of trajectory prediction are no longer limited to easier scenes and trajectories. More and more heterogeneous trajectories with different representation forms, such as 2D or 3D coordinates, 2D or 3D bounding boxes, and even high-dimensional human skeletons, need to be analyzed and forecasted. Among these heterogeneous trajectories, interactions between different elements within a frame of trajectory, which we call the "Dimension-Wise Interactions", would be more complex and challenging. However, most previous approaches focus mainly on a specific form of trajectories, which means these methods could not be used to forecast heterogeneous trajectories, not to mention the dimension-wise interaction. Besides, previous methods mostly treat trajectory prediction as a normal time sequence generation task, indicating that these methods may require more work to directly analyze agents' behaviors and social interactions at different temporal scales. In this paper, we bring a new "view" for trajectory prediction to model and forecast trajectories hierarchically according to different frequency portions from the spectral domain to learn to forecast trajectories by considering their frequency responses. Moreover, we try to expand the current trajectory prediction task by introducing the dimension M from "another view", thus extending its application scenarios to heterogeneous trajectories vertically. Finally, we adopt the bilinear structure to fuse two factors, including the frequency response and the dimension-wise interaction, to forecast heterogeneous trajectories via spectrums hierarchically in a generic way. Experiments show that the proposed model outperforms most state-of-the-art methods on ETH-UCY benchmark, Stanford Drone Dataset and nuScenes with heterogeneous trajectories, including 2D coordinates, 2D and 3D bounding boxes. ! * Equal contribution. Codes are available at https://github.com/cocoon2wong/E-Vertical.

show abstract

ScePT: Scene-consistent, Policy-based Trajectory Predictions for Planning

Cited by 33 publications

References 32 publications

GATraj: A Graph- and Attention-based Multi-Agent Trajectory Prediction Model

GATraj: A Graph- and Attention-based Multi-Agent Trajectory Prediction Model

Pishgu: Universal Path Prediction Architecture through Graph Isomorphism and Attentive Convolution

View Vertically: A Hierarchical Network for Trajectory Prediction via Fourier Spectrums

Contact Info

Product

Resources

About