Summary of Wttfnet: a Weather-time-trajectory Fusion Network For Pedestrian Trajectory Prediction in Urban Complex, by Ho Chun Wu et al.
WTTFNet: A Weather-Time-Trajectory Fusion Network for Pedestrian Trajectory Prediction in Urban Complex
by Ho Chun Wu, Esther Hoi Shan Lau, Paul Yuen, Kevin Hung, John Kwok Tai Chui, Andrew Kwok Fai Lui
First submitted to arxiv on: 29 May 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper proposes a novel approach called Weather-Time-Trajectory Fusion Network (WTTFNet) that improves the performance of baseline deep neural network architecture for pedestrian trajectory modelling in urban complexes. The WTTFNet incorporates weather and time-of-day information as an embedding structure, using a gate multimodal unit to fuse multimodal information and deep representation of trajectories. A joint loss function based on focal loss is used to co-optimize both the deep trajectory features and final classifier, aiming to improve accuracy in predicting pedestrians’ intended destinations and trajectories under possible scenarios of class imbalances. Experimental results on the Osaka Asia and Pacific Trade Center (ATC) dataset show a 23.67% increase in classification accuracy, 9.16% reduction of average displacement error, and 7.07% reduction of final displacement error compared to state-of-the-art algorithms. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper creates a new way to predict where people are going in an urban area based on weather and time. It’s like trying to guess what someone will do next based on what they’re wearing and when they’re doing it! The authors came up with a special tool called the Weather-Time-Trajectory Fusion Network that takes into account things like rain or sunshine, morning or evening, and even what shops people might be going to. They tested this tool on real data from a big shopping center in Japan and found that it was way better than other tools at guessing where people were headed. This could help make cities safer and more efficient for everyone! |
Keywords
» Artificial intelligence » Classification » Embedding » Loss function » Neural network
