English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Predicting distant future trajectories of agents in a dynamic scene is not aneasy problem because the future trajectory of an agent is affected by not onlyhis/her past trajectory but also the scene contexts. To tackle this problem, wepropose a model based on recurrent neural networks (RNNs) and a novel methodfor training the model. The proposed model is based on an encoder-decoderarchitecture where the encoder encodes inputs (past trajectories and scenecontext information) while the decoder produces a trajectory from the contextvector given by the encoder. We train the networks of the proposed model toproduce a future trajectory, which is the closest to the true trajectory, whilemaximizing a reward from a reward function. The reward function is also trainedat the same time to maximize the margin between the rewards from theground-truth trajectory and its estimate. The reward function plays the role ofa regularizer for the proposed model so the trained networks are able to betterutilize the scene context information for the prediction task. We evaluated theproposed model on several public datasets. Experimental results show that theprediction performance of the proposed model is much improved by theregularization, which outperforms the-state-of-the-arts in terms of accuracy.The implementation codes are available athttps://github.com/d1024choi/traj-pred-irl/.

Regularizing Neural Networks for Future Trajectory Prediction via Inverse Reinforcement Learning Framework