Spatial-Temporal Gaussian Scale Mixture Modeling for Foreground Estimation
Authors: Qian Ning, Weisheng Dong, Fangfang Wu, Jinjian Wu, Jie Lin, Guangming Shi11791-11798
AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimental results on real video datasets show that the proposed method performs comparable or even better than current state-of-the-art background subtraction methods. ... The proposed methods were implemented under the Matlab platform. The F-measure is used to evaluate the performance of the proposed methods, defined as ... 5.1 Ablation study To verify the effects of the proposed temporal and spatial regularization terms, we implemented three variants of the proposed method... |
| Researcher Affiliation | Academia | 1State Key Laboratory on Integrated Services Networks (ISN) 2School of Artificial Intelligence, Xidian University, Xi an 710071, China |
| Pseudocode | Yes | Algorithm 1 Proposed STGSM model based Foreground Estimation Method |
| Open Source Code | No | The paper does not provide concrete access to source code for the methodology described. |
| Open Datasets | Yes | The performances of the proposed methods are evaluated on two benchmark datasets, i.e., the perception test image sequences (PTIS) (Li et al. 2004) and the change detection (CD) 2012 (Goyette et al. 2012) datasets, containing 40 videos. |
| Dataset Splits | No | The paper does not provide specific dataset split information (exact percentages, sample counts, citations to predefined splits, or detailed splitting methodology) needed to reproduce the data partitioning into training, validation, and test sets. |
| Hardware Specification | Yes | The proposed methods were implemented with Matlab platform on an Intel Core i7-8700K 3.4GHz CPU, and can be accelerated with parallel computation technique for real-time applications. |
| Software Dependencies | No | The proposed methods were implemented under the Matlab platform. (No specific version number for Matlab or other software dependencies are provided). |
| Experiment Setup | Yes | The major parameters of the proposed methods were empirically set as, r = 15, η = 400/N, σ2 n = 1.05 × 10^-3, and δ = 0.01, which are the same for all test sequences. |