DRNet: Double Recalibration Network for Few-Shot Semantic Segmentation

Gao, Guangyu; Fang, Zujie; Han, Cen; Wei, Yunchao; Liu, Chi Harold; Yan, Shuicheng

doi:10.1109/tip.2022.3215905

Cited by 11 publications

(2 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…DGPNet [49] proposed to fit a set of gaussian process models based on support images and their corresponding masks, to compute the mean and variance of the foreground feature distribution. The prototype-based approaches [52]- [57] cluster the features into one or more prototypes representing particular semantics and compare prototypes with query features to discover the target area. PANet [55] presented a prototype alignment regularization between support and query for class-specific knowledge representation to achieve better generalization performance.…”

Section: Few-shot Semantic Segmentationmentioning

confidence: 99%

“…PPNet [54] proposed a prototype representation method that decomposed the overall class representation into partial prototype awareness to capture enriched and fine-grained feature representations. DRNet [52] aims to solve the intra-class variance of the unseen class through two recalibration modules, which explore latent regions for query images. IMPT [57] suggests using an intermediate prototype to extract class information from deterministic support features and adaptive query features.…”

Section: Few-shot Semantic Segmentationmentioning

confidence: 99%

See 1 more Smart Citation

Holistic Prototype Attention Network for Few-Shot Video Object Segmentation

Tang

Chen

Jiang

et al. 2024

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

Few-shot video object segmentation (FSVOS) aims to segment dynamic objects of unseen classes by resorting to a small set of support images that contain pixel-level object annotations. Existing methods have demonstrated that the domain agent-based attention mechanism is effective in FSVOS by learning the correlation between support images and query frames. However, the agent frame contains redundant pixel information and background noise, resulting in inferior segmentation performance. Moreover, existing methods tend to ignore inter-frame correlations in query videos. To alleviate the above dilemma, we propose a holistic prototype attention network (HPAN) for advancing FSVOS. Specifically, HPAN introduces a prototype graph attention module (PGAM) and a bidirectional prototype attention module (BPAM), transferring informative knowledge from seen to unseen classes. PGAM generates local prototypes from all foreground features and then utilizes their internal correlations to enhance the representation of the holistic prototypes. BPAM exploits the holistic information from support images and video frames by fusing co-attention and self-attention to achieve support-query semantic consistency and inner-frame temporal consistency. Extensive experiments on YouTube-FSVOS have been provided to demonstrate the effectiveness and superiority of our proposed HPAN method. Our source code and models are available anonymously at https: //github.com/NUST-Machine-Intelligence-Laboratory/HPAN.

show abstract