Recognizing Complex Activities by a Probabilistic Interval-Based Model
Authors: Li Liu, Li Cheng, Ye Liu, Yongpo Jia, David Rosenblum
AAAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Empirical evaluations on benchmark datasets suggest our approach significantly outperforms the state-of-the-art methods. |
| Researcher Affiliation | Academia | 1School of Computing, National University of Singapore, Singapore 117417 2Bioinformatics Institute, A*STAR, Singapore 138671 |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper provides a link to 'supplemental material' (https://drive.google.com/folderview?id=0B20pV67EQA7Q3NHR3hWVG85MFE) but does not explicitly state that the source code for the described methodology is available at this link. |
| Open Datasets | Yes | Three complex activity recognition datasets are considered in our experiments. OSUPEL dataset (Brendel, Fern, and Todorovic 2011). Opportunity dataset (Roggen et al. 2010). Composable activities dataset (CAD14) (Lillo, Soto, and Niebles 2014). |
| Dataset Splits | Yes | Table 3 shows the averaged accuracy results over 5-fold cross-validations. |
| Hardware Specification | No | The paper mentions runtime comparisons but does not specify any hardware details such as CPU, GPU models, or memory used for the experiments. |
| Software Dependencies | No | The paper does not provide specific version numbers for any software dependencies (e.g., programming languages, libraries, or frameworks) used in the experiments. |
| Experiment Setup | Yes | Besides, the hyperparameters α and β are generally unknown before the start of Gibbs sampling and therefore need to be estimated. In our experiment, we used the convergent method (Minka 2000) that iteratively updates these hyperparameters by approximately estimating the objective maximum likelihood function values. Besides, a small smoothing constant S (S = 0.00001) is introduced to avoid the numerical issue of division by zero. |