Salient Object Detection via Objectness Proposals

Authors: Tam Nguyen

AAAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental The widely used ROC curves of our system to other works in MSRA-1000 dataset (Achanta et al. 2009) are shown in Figure 2. Our method outperforms other state-of-the-art baselines (AIM (Bruce and Tsotsos 2005), FT (Achanta et al. 2009), GC (Cheng et al. 2013), IT (Itti, Koch, and Niebur 1998), LC (Zhai and Shah 2006), SF (Perazzi et al. 2012), SR (Hou and Zhang 2007)) In addition, we also compare the average running time of our approach to the currently best performing methods on the benchmark images.
Researcher Affiliation Academia Tam V. Nguyen Department for Technology, Innovation and Enterprise Singapore Polytechnic, Singapore
Pseudocode No The paper describes the algorithm in prose and provides an overview figure, but does not include structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide concrete access to source code for the methodology described in this paper.
Open Datasets Yes BING proposal generator is trained with VOC2007 dataset (Everingham et al. 2010) same as in (Cheng et al. 2014). The widely used ROC curves of our system to other works in MSRA-1000 dataset (Achanta et al. 2009) are shown in Figure 2.
Dataset Splits No The paper does not provide specific dataset split information (exact percentages, sample counts, or detailed splitting methodology) for training, validation, or testing.
Hardware Specification Yes The average time is taken by each method on a PC with Intel i7 3.3 GHz CPU and 8GB RAM.
Software Dependencies No The paper mentions software like MATLAB, C++, SLIC, and BING, but does not provide specific version numbers for any of these components.
Experiment Setup Yes We set the number of superpixels as 100 as a trade-off between the fine oversegmentation and the processing time.