相关参考文献
注意：仅列出部分参考文献，下载原文获取全部文献信息。
Article
Engineering, Electrical & Electronic
I-Hsi Kao et al.
Summary: This study conducted a comparative experiment on machine learning-based pedestrian trajectory prediction and found that incorporating social and posture features can improve prediction accuracy. Deep learning methods perform better in handling complex features, especially the 3D CNN model that combines social and posture features.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Computer Science, Artificial Intelligence
Liangyu Chai et al.
Summary: This paper introduces a research problem aimed at generating diverse and continuous crowd videos, proposing a deep network architecture specifically designed for crowd video generation. Experimental results show the method is effective and the generated videos can be utilized for various crowd analysis tasks.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Pei Xu et al.
Summary: This paper proposes a novel approach called SocialVAE for human trajectory prediction. SocialVAE utilizes stochastic recurrent neural networks, social attention mechanism, and backward posterior approximation to better predict pedestrian navigation strategies, achieving better performance than existing methods on various benchmark datasets.
COMPUTER VISION - ECCV 2022, PT IV
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Tianpei Gu et al.
Summary: This paper presents a new framework for trajectory prediction based on the reverse process of motion indeterminacy diffusion. By learning a parameterized Markov chain from observed trajectories, the degree of indeterminacy can be controlled and the diversity and determinacy of predictions can be balanced. Extensive experiments demonstrate the superiority of this method.
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2022)
(2022)
Article
Engineering, Electrical & Electronic
Xiaobo Chen et al.
Summary: In this article, an intention-aware non-autoregressive Transformer model is proposed for accurately predicting multimodal vehicle trajectory. The model utilizes social attention learning and temporal attention learning to capture social interaction and temporal dependency, and cross-attention learning to extract features and make future predictions.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Engineering, Electrical & Electronic
Wendong Xiao et al.
Summary: This article introduces a novel actor-critic architecture for learning optimal navigation policy. It reshapes the reward function and introduces collision penalty for collision avoidance and obtaining measurement information from visual observation. Expert trajectories are used to generate subgoals and an observation-action consistency model is introduced for human-aware navigation policies. The proposed method achieves better performance and faster convergence speed compared to existing approaches, with improved generalization capacity.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Automation & Control Systems
T. Ran et al.
Summary: The paper introduces a vision-based navigation approach using deep learning techniques and shallow CNN to improve scene classification accuracy and efficiency, as well as introducing a combination of AWC algorithm and RC control to enhance robot motion performance.
Article
Robotics
Yu Yao et al.
Summary: This letter introduces BiTraP, a goal-conditioned bidirectional multi-modal trajectory prediction method based on CVAE, which accurately predicts pedestrian trajectory goals and improves long-term prediction accuracy. Extensive experiments show BiTraP's superior performance across different scenarios, outperforming state-of-the-art methods by 10-50% in accuracy.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Karttikeya Mangalam et al.
Summary: Human trajectory forecasting involves both known and unknown sources of uncertainty, which researchers have proposed to decompose into epistemic and aleatoric sources. They have introduced a novel long-term trajectory forecasting setting and a Y-net model that significantly improves performance compared to previous works in long prediction horizons.
2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021)
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Parth Kothari et al.
Summary: This study addresses the challenge of human trajectory forecasting by combining neural networks and discrete choice models to improve prediction accuracy and interpretability.
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Chengxin Wang et al.
Summary: The paper introduces a novel CNN-based spatial-temporal graph framework called GraphTCN, which efficiently and accurately predicts the future paths of an agent's neighbors by simulating the spatial and temporal interactions. The model computes spatial and temporal modeling within each local time window, allowing for parallel execution and higher efficiency.
2021 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION WACV 2021
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Francesco Giuliari et al.
Summary: Recent successes in forecasting people motion have largely been based on LSTM models, but this study proposes a new method using Transformer Networks for trajectory forecasting, which transitions from sequential processing to attention-based memory mechanisms. Results show that using Transformer models can achieve better performance in the most challenging trajectory forecasting benchmarks.
2020 25TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR)
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Yue Song et al.
Summary: Understanding human behaviors in crowded scenarios requires analyzing both the position of the subjects in space and the scene context. Existing approaches rely on motion history and interactions among people, while the proposed model uses coherent group clustering and a global attention mechanism to address motion prediction. The attentive group-aware GAN outperformed state-of-the-art models on benchmark datasets and generated socially-acceptable trajectories.
2020 25TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR)
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Joey Hong et al.
2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019)
(2019)
Proceedings Paper
Computer Science, Artificial Intelligence
Shuai Yi et al.
COMPUTER VISION - ECCV 2016, PT I
(2016)
Article
Computer Science, Software Engineering
Alon Lerner et al.
COMPUTER GRAPHICS FORUM
(2007)
