FOCUS: Towards Universal Foreground Segmentation
Authors: Zuyao You, Lingyu Kong, Lingchen Meng, Zuxuan Wu
AAAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We conduct extensive experiments on a total of 13 datasets across 5 tasks, and the results demonstrate that FOCUS consistently outperforms the state-of-the-art task-specific models on most metrics. |
| Researcher Affiliation | Academia | 1Shanghai Key Lab of Intell. Info. Processing, School of CS, Fudan University 2Shanghai Collaborative Innovation Center of Intelligent Visual Computing EMAIL, EMAIL |
| Pseudocode | No | The paper describes the proposed FOCUS framework, including its components like the backbone, edge enhancer, feature decoder, and CLIP refiner, through textual descriptions and mathematical formulations. However, it does not include any explicitly labeled pseudocode or algorithm blocks. |
| Open Source Code | Yes | Code https://github.com/geshang777/FOCUS/ |
| Open Datasets | Yes | For COD, we follow (Fan et al. 2021; Zheng et al. 2024a), training FOCUS on the combination of CAMO-TR (Le et al. 2019) and COD10K-TR (Fan et al. 2020) and evaluating on CAMO-TE, COD10K-TE, CHAMELEON (Skurowski et al. 2018) and NC4K (Lv et al. 2021). For SOD task, we follow (Wang et al. 2023), using DUTSTR (Wang et al. 2017) as training dataset without extra data, evaluating our model on DUTS-TE, DUT-OMRON (Yang et al. 2013), HKU-IS (Li and Yu 2015), ECSSD (Shi et al. 2015) and PACAL-S (Li et al. 2014) respectively. For SD, We use ISTD (Wang, Li, and Yang 2018) as our training and evaluation dataset. For DBD, following previous work (Zhao et al. 2018), we use the combination of CUHK (Shi, Xu, and Jia 2014) and DUT (Zhao et al. 2018) as training dataset, and the remaining 100 images in CUHK and 500 images in DUT for testing. Following (Wang et al. 2022a), we use CASIA-2.0 (Dong, Wang, and Tan 2013) as the training dataset and evaluate on CASIA-1.0. |
| Dataset Splits | Yes | For COD, we follow (Fan et al. 2021; Zheng et al. 2024a), training FOCUS on the combination of CAMO-TR (Le et al. 2019) and COD10K-TR (Fan et al. 2020) and evaluating on CAMO-TE, COD10K-TE, CHAMELEON (Skurowski et al. 2018) and NC4K (Lv et al. 2021). For SOD task, we follow (Wang et al. 2023), using DUTSTR (Wang et al. 2017) as training dataset without extra data, evaluating our model on DUTS-TE, DUT-OMRON (Yang et al. 2013), HKU-IS (Li and Yu 2015), ECSSD (Shi et al. 2015) and PACAL-S (Li et al. 2014) respectively. For SD, We use ISTD (Wang, Li, and Yang 2018) as our training and evaluation dataset. For DBD, following previous work (Zhao et al. 2018), we use the combination of CUHK (Shi, Xu, and Jia 2014) and DUT (Zhao et al. 2018) as training dataset, and the remaining 100 images in CUHK and 500 images in DUT for testing. Following (Wang et al. 2022a), we use CASIA-2.0 (Dong, Wang, and Tan 2013) as the training dataset and evaluate on CASIA-1.0. |
| Hardware Specification | Yes | We use batch size 8 for all experiments and 2 NVIDIA A6000 GPUs with 48G memory. |
| Software Dependencies | Yes | Our framework is implemented using Py Torch 2.1.1 (Paszke et al. 2019). |
| Experiment Setup | Yes | We use batch size 8 for all experiments... The FOCUS is trained on each training dataset with the size of 512 512 for 20,000 iterations on average with Adam W optimizer (Loshchilov and Hutter 2017). The initial learning rate is set to 10 5 with a weight decay of 0.05 to regularize the model. The L2 norm is used for gradient clipping, and the maximum allowed value for gradients is set to 0.01. |