How Unlabeled Web Videos Help Complex Event Detection?

Authors: Huan Liu, Qinghua Zheng, Minnan Luo, Dingwen Zhang, Xiaojun Chang, Cheng Deng

IJCAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiment results over standard datasets of TRECVID MEDTest 2013 and TRECVID MEDTest 2014 demonstrate the effectiveness and superiority of the proposed framework on complex event detection.
Researcher Affiliation Academia 1MOEKLINNS Lab, Department of Computer Science, Xi an Jiaotong University, Shaanxi, China; 2School of Automation, Northwestern Polytechnical University, Shaanxi, China; 3School of Computer Science, Carnegie Mellon University, PA, USA; 4School of Electronic Engineering, Xidian University, Shaanxi, China
Pseudocode Yes Algorithm 1 Alternating algorithm for problem (1)
Open Source Code No The paper does not contain any explicit statement about releasing source code for the described methodology or a link to a code repository.
Open Datasets Yes We evaluate on two large scale real-world datasets: the TRECVID MEDTest 2013 and the TRECVID MEDTest 2014... We use the Yahoo Flickr Creative Commons 100 Million Dataset (YFCC100M) [Thomee et al., 2016] as the unlabeled web videos in the experiments. ... 3http://nist.gov/itl/iad/mig/med13.cfm 4http://nist.gov/itl/iad/mig/med14.cfm
Dataset Splits No The paper mentions 'cross-validated the regularization parameters' but does not specify the explicit training, validation, and test dataset splits used for its experiments.
Hardware Specification No The paper does not provide any specific details about the hardware (e.g., GPU/CPU models, memory) used for running the experiments.
Software Dependencies No The paper mentions using 'CNN features... VGG' and 'VLAD encoding' but does not provide specific software names with version numbers for reproducibility.
Experiment Setup Yes We cross-validated the regularization parameters in the range of {0.01, 0.1, 1, 10, 100}. We set p = 0.8 and q = 1.2 in our experiments to achieve the best performance.