Ensemble-Based Tracking: Aggregating Crowdsourced Structured Time Series Data
Authors: Naiyan Wang, Dit-Yan Yeung
ICML 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Using the largest open benchmark for visual tracking, we empirically compare two ensemble methods constructed from five state-of-the-art trackers with the individual trackers. The promising experimental results provide empirical evidence for our ensemble approach to get the best of all worlds . also 6. Experiments To facilitate objective comparison, we use a recently released benchmark (Wu et al., 2013) in our experiments. |
| Researcher Affiliation | Academia | Naiyan Wang WINSTY@GMAIL.COM Dit-Yan Yeung DYYEUNG@CSE.UST.HK Department of Computer Science and Engineering, Hong Kong Univeristy of Science and Technology Clear Water Bay, Hong Kong |
| Pseudocode | Yes | Algorithm 1 Conditional Particle Filter Algorithm |
| Open Source Code | Yes | The implementation of EBT and SC-EBT can be found on the project page: http://winsty.net/ ebt.html. |
| Open Datasets | Yes | To facilitate objective comparison, we use a recently released benchmark (Wu et al., 2013) in our experiments. It is currently the largest open benchmark for visual tracking, which comprises 51 video sequences covering 11 challenging aspects of visual tracking. |
| Dataset Splits | No | The paper references a benchmark dataset (Wu et al., 2013) but does not specify how the dataset was partitioned into explicit training, validation, and test splits for the experiments described in this paper. |
| Hardware Specification | No | The paper does not provide specific details about the hardware used for running experiments, such as CPU/GPU models, memory, or cloud instance types. |
| Software Dependencies | No | The paper mentions that a web service facilitates communication between trackers implemented in different languages and operating systems, and lists specific trackers used (ASLA, Struck, DLT, CSK, LSST), but it does not provide specific version numbers for any software components or libraries. |
| Experiment Setup | Yes | We set M = 50, N = 400 for the particle filter, k = 0.1, α = 2, a = 0.1 for the model. For failure detection, we set p = 10 and θ = 0.8T and 0.9T in EBT and SC-EBT, respectively. For the Gaussian transition probability distributions for horizontal translation, vertical translation, scale and aspect ratio, their standard deviations are 4, 4, 0.01, and 0.001, respectively. |