A Complete Variational Tracker

Authors: Ryan D Turner, Steven Bottone, Bhargav Avasarala

NeurIPS 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We demonstrate the applicability of our method on radar tracking and computer vision problems. We use the VS-PETS 2003 soccer player data set as a real data example to validate our method.
Researcher Affiliation Industry Ryan Turner Northrop Grumman Corp. ryan.turner@ngc.com Steven Bottone Northrop Grumman Corp. steven.bottone@ngc.com Bhargav Avasarala Northrop Grumman Corp. bhargav.avasarala@ngc.com
Pseudocode No The paper describes algorithms such as 'variational algorithm' and 'loopy belief propagation', and provides mathematical derivations of steps, but it does not include a dedicated pseudocode block or a clearly labeled algorithm section.
Open Source Code No The paper does not contain any explicit statement about making the source code available or provide a link to a code repository.
Open Datasets Yes We borrow the radar tracking example of the OMGP paper [18]." and "We use the VS-PETS 2003 soccer player data set as a real data example to validate our method. ... Soccer data source: http://www.cvg.rdg.ac.uk/slides/pets.html.
Dataset Splits No The paper states, 'The parameters for the NCV, R, PD, λ, and the track meta-state parameters were trained by optimizing the variational lower bound Lβ on the first 1000 frames...' and 'We split the remainder of the data into 70 sequences of K = 20 frames for a test set.' This indicates training and testing sets, but no explicit validation set.
Hardware Specification No The paper does not provide any specific hardware details such as GPU/CPU models, processor types, or memory amounts used for running its experiments.
Software Dependencies No The paper does not provide specific ancillary software details with version numbers, such as library or solver names, that are needed to replicate the experiment.
Experiment Setup Yes We borrow the radar tracking example... We have made the example more realistic by adding clutter λ = 8 and missed detections PD = 0.5". For the soccer example: "The parameters for the NCV, R, PD, λ, and the track meta-state parameters were trained by optimizing the variational lower bound Lβ on the first 1000 frames".