Where will they go? Predicting fine-grained adversarial multi-agent motion using conditional variational autoencoders
Simultaneously and accurately forecasting the behavior of many interacting agents is imperative for computer vision applications to be widely deployed (e.g., autonomous vehicles, security, surveillance, sports). In this paper, we present a technique using conditional variational autoencoder which learns a model that "personalizes" prediction to individual agent behavior within a group representation. Given the volume of data available and its adversarial nature, we focus on the sport of basketball and show that our approach efficiently predicts context-specific agent motions. We find that our model generates results that are three times as accurate as previous state of the art approaches (5.74 ft vs. 17.95 ft).
© Copyright 2018 Computer Vision - ECCV 2018. Lecture Notes in Computer Science: 15th European Conference, Munich, Germany, September 8-14, 2018. Julkaistu Tekijä Springer. Kaikki oikeudet pidätetään.
Aiheet: | teknologia matemaattis-looginen malli koripallo urheilupeli mallintaminen tietokone ennuste |
---|---|
Aihealueet: | tekniset ja luonnontieteet urheilukilpailut |
Tagging: | Autoencoder |
DOI: | 10.1007/978-3-030-01252-6_45 |
Julkaisussa: | Computer Vision - ECCV 2018. Lecture Notes in Computer Science: 15th European Conference, Munich, Germany, September 8-14, 2018 |
Toimittajat: | V. Ferrari, M. Hebert, C. Sminchisescu, Y. Weiss |
Julkaistu: |
Cham
Springer
2018
|
Sivuja: | 732-747 |
Julkaisutyypit: | kongressin muistiinpanot |
Kieli: | englanti (kieli) |
Taso: | kehittynyt |