Multi-view Feature Augmentation with Adaptive Class Activation Mapping

Authors: Xiang Gao, Yingjie Tian, Zhiquan Qi

IJCAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments demonstrate consistent and noticeable performance gains achieved by our multi-view feature augmentation module.
Researcher Affiliation Academia 1School of Computer Science and Technology, University of Chinese Academy of Sciences 2Research Center on Fictitious Economy and Data Science, Chinese Academy of Sciences 3Key Laboratory of Big Data Mining and Knowledge Management, Chinese Academy of Sciences gaoxiang181@mails.ucas.ac.cn, tyj@ucas.ac.cn, qizhiquan@foxmail.com
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide concrete access to source code for the methodology described.
Open Datasets Yes We evaluate our method on the following object, scene, and action classification datasets: Caltech101, Imagenette, Corel5k, Scene15, MIT Indoor67, Stanford Action40, UIUC Event8, UCMLU, RSSCN7, and AID. The details of these datasets are described in supplementary materials.
Dataset Splits No The paper mentions 'validation accuracy' and evaluates models on 'validation sets' (e.g., Imagenette validation set) but does not provide specific dataset split information (percentages, sample counts, or detailed splitting methodology) for reproducibility.
Hardware Specification Yes Our model is implemented with Tensor Flow and run on a single Ge Force GTX 1080 Ti GPU.
Software Dependencies No The paper mentions 'Tensor Flow' as the implementation framework but does not provide a specific version number or list other software dependencies with their versions.
Experiment Setup Yes We use the Adam optimizer with β1 = 0.9 and β2 = 0.999. The initial learning rate is lr = 0.0001, the batch size is N = 64. We set K = 50 and LR = [3, 5, 7, 9] in our MVFea Aug module. We train 20k iterations on Imagenette, Caltech101, AID, and Corel5k, 10k iterations on Scene15, Event8, and RSSCN7, 15k iterations on UCMLU, 30k iterations on Action40, and 40k iterations on Indoor67, making sure that learning converges on all datasets.