Trajectory Prediction On Stanford Drone

Metrics

ADE (8/12) @K=5

FDE(8/12) @K=5

Results

Performance results of various models on this benchmark

Model Name	ADE (8/12) @K=5	FDE(8/12) @K=5	Paper Title	Repository
TNT	12.23	21.16	TNT: Target-driveN Trajectory Prediction	-
Y-Net	-	-	From Goals, Waypoints & Paths To Long Term Human Trajectory Forecasting	-
CGNS	-	-	Conditional Generative Neural System for Probabilistic Trajectory Prediction	-
SoPhie	16.27	29.38	SoPhie: An Attentive GAN for Predicting Paths Compliant to Social and Physical Constraints	-
Multiverse	-	-	The Garden of Forking Paths: Towards Multi-Future Trajectory Prediction	-
DAG-Net	-	-	DAG-Net: Double Attentive Graph Neural Network for Trajectory Forecasting	-
SMEMO	11.64	21.12	SMEMO: Social Memory for Trajectory Forecasting	-
Social-Implicit	-	-	Social-Implicit: Rethinking Trajectory Prediction Evaluation and The Effectiveness of Implicit Maximum Likelihood Estimation	-
DESIRE	19.25	34.05	DESIRE: Distant Future Prediction in Dynamic Scenes with Interacting Agents	-
V^2-Net	-	-	View Vertically: A Hierarchical Network for Trajectory Prediction via Fourier Spectrums	-
Social LSTM	31.19	56.97	Social LSTM: Human Trajectory Prediction in Crowded Spaces	-
MemoNet	-	-	Remember Intentions: Retrospective-Memory-based Trajectory Prediction	-
TDOR	-	-	End-to-End Trajectory Distribution Prediction Based on Occupancy Grid Maps	-
NSP-SFM	-	-	Human Trajectory Prediction via Neural Social Physics	-
Sophie	-	-	SoPhie: An Attentive GAN for Predicting Paths Compliant to Social and Physical Constraints	-
P2TIRL	-	-	Trajectory Forecasts in Unknown Environments Conditioned on Grid-Based Plans	-
CF-VAE	-	-	Conditional Flow Variational Autoencoders for Structured Sequence Prediction	-
PPT	-	-	Progressive Pretext Task Learning for Human Trajectory Prediction	-
EvolveGraph	-	-	EvolveGraph: Multi-Agent Trajectory Prediction with Dynamic Relational Reasoning	-
Social GAN	27.25	41.44	Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networks	-

0 of 24 row(s) selected.