Projective dictionary pair learning for pattern classification
Authors: Shuhang Gu, Lei Zhang, Wangmeng Zuo, Xiangchu Feng
NeurIPS 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We evaluate the proposed DPL method on various visual classification datasets, including two face databases (Extended Yale B [22] and AR [24]), one object categorization database (Caltech101) [25], and one action recognition database (UCF 50 action [26]). These datasets are widely used in previous works [5, 9] to evaluate the DL algorithms. Besides the classification accuracy, we also report the training and testing time of competing algorithms in the experiments. |
| Researcher Affiliation | Academia | Shuhang Gu1, Lei Zhang1, Wangmeng Zuo2, Xiangchu Feng3 1Dept. of Computing, The Hong Kong Polytechnic University, Hong Kong, China 2School of Computer Science and Technology, Harbin Institute of Technology, Harbin, China 3Dept. of Applied Mathematics, Xidian University, Xi an, China |
| Pseudocode | Yes | Algorithm 1 Discriminative synthesis&analysis dictionary pair learning (DPL) Input: Training samples for K classes X = [X1, X2, . . . , XK], parameter λ, τ, m; 1: Initialize D(0) and P(0) as random matrixes with unit Frobenious norm, t = 0; 2: while not converge do 3: t t + 1; 4: for i=1:K do 5: Update A(t) k by (8); 6: Update P(t) k by (10); 7: Update D(t) k by (12); 8: end for 9: end while Output: Analysis dictionary P, synthesis dictionary D. |
| Open Source Code | No | The paper does not provide any explicit statement about releasing the source code or a link to a code repository for the described methodology. |
| Open Datasets | Yes | We evaluate the proposed DPL method on various visual classification datasets, including two face databases (Extended Yale B [22] and AR [24]), one object categorization database (Caltech101) [25], and one action recognition database (UCF 50 action [26]). |
| Dataset Splits | Yes | We randomly select half of the images per subject for training and the other half for testing. For the AR database, a set of 2,600 images of 50 female and 50 male subjects are extracted. 20 images of each subject are used for training and the remain 6 images are used for testing. For the Caltech101 database...30 samples per category are used for training and the rest are used for testing. We use the action bank features [28] and five-fold data splitting to evaluate our algorithm. In all the experiments, we choose λ and τ by 10-fold cross validation on each dataset. |
| Hardware Specification | Yes | All experiments are run on a desktop PC with 3.5GHz Intel CPU and 8 GB memory. |
| Software Dependencies | No | The paper states that 'All the competing algorithms are implemented in Matlab except for SVM which is implemented in C' but does not provide specific version numbers for these software components or any other libraries. |
| Experiment Setup | Yes | There are three parameters, m, λ and τ in the proposed DPL model. In all the experiments, we choose λ and τ by 10-fold cross validation on each dataset. For all the competing methods, we tune their parameters for the best performance. The parameter τ is set to 0.05 on both the datasets and λ is set to 3e-3 and 5e-3 on the Extended Yale B and AR datasets, respectively. The parameters τ and λ used in our algorithm are 0.05 and 1e-4, respectively. The parameters τ and λ used in our algorithm are both 0.01. |