1 code implementation • 7 Feb 2024 • Hersh Vakharia, Xiaoxiao Du
Previously, we developed a Multiple Instance Multi-Resolution Fusion (MIMRF) framework that addresses label uncertainty for fusion, but it can be slow to train due to the large search space for the fuzzy measures used to integrate sensor data sources.
1 code implementation • 5 Jul 2022 • Asiegbu Miracle Kanu-Asiegbu, Ram Vasudevan, Xiaoxiao Du
We present BiPOCO, a Bi-directional trajectory predictor with POse COnstraints, for detecting anomalous activities of pedestrians in videos.
Ranked #6 on Video Anomaly Detection on HR-Avenue
1 code implementation • 5 Jul 2022 • Asiegbu Miracle Kanu-Asiegbu, Ram Vasudevan, Xiaoxiao Du
Video anomaly detection is a core problem in vision.
1 code implementation • 10 May 2021 • Yu Yao, Ella Atkins, Matthew Johnson Roberson, Ram Vasudevan, Xiaoxiao Du
In this work, we follow the neuroscience and psychological literature to define pedestrian crossing behavior as a combination of an unobserved inner will (a probabilistic representation of binary intent of crossing vs. not crossing) and a set of multi-class actions (e. g., walking, standing, etc.).
1 code implementation • 29 Jul 2020 • Yu Yao, Ella Atkins, Matthew Johnson-Roberson, Ram Vasudevan, Xiaoxiao Du
BiTraP estimates the goal (end-point) of trajectories and introduces a novel bi-directional decoder to improve longer-term trajectory prediction accuracy.
Ranked #2 on Trajectory Prediction on JAAD
1 code implementation • 5 Mar 2019 • Cyrus Anderson, Xiaoxiao Du, Ram Vasudevan, Matthew Johnson-Roberson
Our work demonstrates the effectiveness and potential of using simulation as a substitution for human annotation efforts to train high-performing prediction algorithms such as the DNNs.
Robotics
no code implementations • 11 Sep 2018 • Xiaoxiao Du, Ram Vasudevan, Matthew Johnson-Roberson
In applications such as autonomous driving, it is important to understand, infer, and anticipate the intention and future behavior of pedestrians.
3 code implementations • 2 May 2018 • Xiaoxiao Du, Alina Zare
It is valuable to fuse outputs from multiple sensors to boost overall performance.
2 code implementations • 11 Mar 2018 • Xiaoxiao Du, Alina Zare
In classifier (or regression) fusion the aim is to combine the outputs of several algorithms to boost overall performance.