DensePeds uses Front-RVO motion prediction and Mask R-CNN sparse features to track people in crowds denser than 2 per square meter, running 4.5 times faster than prior methods and improving accuracy by 2.6% on dense datasets.
Recurrent Autoregressive Networks for Online Multi-Object Tracking
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
The main challenge of online multi-object tracking is to reliably associate object trajectories with detections in each video frame based on their tracking history. In this work, we propose the Recurrent Autoregressive Network (RAN), a temporal generative modeling framework to characterize the appearance and motion dynamics of multiple objects over time. The RAN couples an external memory and an internal memory. The external memory explicitly stores previous inputs of each trajectory in a time window, while the internal memory learns to summarize long-term tracking history and associate detections by processing the external memory. We conduct experiments on the MOT 2015 and 2016 datasets to demonstrate the robustness of our tracking method in highly crowded and occluded scenes. Our method achieves top-ranked results on the two benchmarks.
fields
cs.RO 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
DensePeds: Pedestrian Tracking in Dense Crowds Using Front-RVO and Sparse Features
DensePeds uses Front-RVO motion prediction and Mask R-CNN sparse features to track people in crowds denser than 2 per square meter, running 4.5 times faster than prior methods and improving accuracy by 2.6% on dense datasets.