Deep Joint Task Learning for Generic Object Extraction
Authors: Xiaolong Wang, Liliang Zhang, Liang Lin, Zhujin Liang, Wangmeng Zuo
NeurIPS 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments suggest that our framework significantly outperforms other state-of-the-art approaches in both accuracy and efficiency (e.g. 1000 times faster than competing approaches). We validate our approach on the Saliency dataset [9, 8] and a more challenging dataset newly collected by us, namely Object Extraction(OE) dataset1. |
| Researcher Affiliation | Academia | 1Sun Yat-sen University, Guangzhou 510006, China 2School of Computer Science and Technology, Harbin Institute of Technology, China 3SYSU-CMU Shunde International Joint Research Institute, Shunde, China 4The Robotics Institute, Carnegie Mellon University, Pittsburgh, U.S. |
| Pseudocode | No | The paper describes an 'EM-type algorithm' and illustrates it with Figure 3, but does not provide structured pseudocode or an algorithm block. |
| Open Source Code | No | The paper does not include an explicit statement about releasing source code or a direct link to a code repository for the described methodology. |
| Open Datasets | Yes | We validate our approach on the Saliency dataset [9, 8] and a more challenging dataset newly collected by us, namely Object Extraction(OE) dataset1. ... The Saliency dataset is a combination of THUR15000 [8] and THUS10000 [9] datasets... We select the images from the PASCAL [15], i Coseg [3], Internet [28] datasets as well as other data (most of them are about people and clothes) from the web. |
| Dataset Splits | No | The paper specifies training and testing splits (e.g., '14233 images for training and 2000 images for testing' and '8230 samples are randomly selected for training and the remaining 1953 ones are applied in testing'), but does not explicitly mention a validation split. |
| Hardware Specification | Yes | The experiments are performed on a desktop with an Intel I7 3.7GHz CPU, 16GB RAM and GTX TITAN GPU. |
| Software Dependencies | No | The paper does not provide specific software dependencies with version numbers. |
| Experiment Setup | Yes | The domain of each element in the 4-dimension latent variable vector Li is set to [ 10, 5, 0, 5, 10], thus there are 54 = 625 possible proposals for each Li. We set the number of MCMC sampling moves as K = 20 during searching. The learning rate is ϵ1 = 1.0 10 6 for the segmentation network and ϵ2 = 1.0 10 8 for the localization network. For testing, as each pixelwise output of our method is well discriminated to the number around 1 or 0, we simply classify it as foreground or background by setting a threshold 0.5. |