当前位置:网站首页>Target driven long term trajectory prediction (CS CV)

Target driven long term trajectory prediction (CS CV)

2020-12-07 19:16:51 Ling Qian

With the application of powerful sequence modeling and rich environment feature extraction technology , Great progress has been made in the prediction of human short-term trajectory . However , Long term forecasting remains a major challenge to current methods , Because errors can accumulate in the process . in fact , Consistent and stable predictions to the end of the trajectory essentially require a more in-depth analysis of the overall structure of the trajectory , This has something to do with the pedestrian's intention to the destination of the journey . In this work , We suggest that a hypothetical process be established to determine pedestrian goals , And the long-term impact on this process . We design a target driven trajectory prediction model —— A two channel neural network to implement this intuition . The two channels of the network play their respective roles , And cooperate with each other , To generate the trajectory of the future . And traditional goal-based 、 Planning based approaches are different , The model architecture is designed to generalize patterns and work in different scenarios with arbitrary geometric and semantic structures . The model in various cases , Especially in the larger prediction range , Its performance is superior to the prior art . This result proves the effectiveness of adaptive structured representation of visual and geometric features in human behavior analysis .

Original title :Goal-driven Long-Term Trajectory Prediction

original text :The prediction of humans' short-term trajectories has advanced significantly with the use of powerful sequential modeling and rich environment feature extraction. However, long-term prediction is still a major challenge for the current methods as the errors could accumulate along the way. Indeed, consistent and stable prediction far to the end of a trajectory inherently requires deeper analysis into the overall structure of that trajectory, which is related to the pedestrian's intention on the destination of the journey. In this work, we propose to model a hypothetical process that determines pedestrians' goals and the impact of such process on long-term future trajectories. We design Goal-driven Trajectory Prediction model - a dual-channel neural network that realizes such intuition. The two channels of the network take their dedicated roles and collaborate to generate future trajectories. Different than conventional goal-conditioned, planning-based methods, the model architecture is designed to generalize the patterns and work across different scenes with arbitrary geometrical and semantic structures. The model is shown to outperform the state-of-the-art in various settings, especially in large prediction horizons. This result is another evidence for the effectiveness of adaptive structured representation of visual and geometrical features in human behavior analysis.

Original author :Hung Tran, Vuong Le, Truyen Tran

Original address :https://arxiv.org/abs/2011.02751

Original statement , This article is authorized by the author + Community publication , Unauthorized , Shall not be reproduced .

If there is any infringement , Please contact the yunjia_community@tencent.com Delete .

版权声明
本文为[Ling Qian]所创,转载请带上原文链接,感谢
https://chowdera.com/2020/11/20201112182545435e.html