Learnable Prompt For Few-shot Semantic Segmentation In Remote Sensing Domain

Abstract

Few-shot segmentation is a task to segment objects or regions of novel classes within an image given only a few annotated examples. In the generalized setting, the task extends to segment both the base and the novel classes. The main challenge is how to train the model such that the addition of novel classes does not hurt the base classes performance, also known as catastrophic forgetting. To mitigate this issue, we use SegGPT as our base model and train it on the base classes. Then, we use separate learnable prompts to handle predictions for each novel class. To handle various object sizes which typically present in remote sensing domain, we perform patch-based prediction. To address the discontinuities along patch boundaries, we propose a patch-and-stitch technique by re-framing the problem as an image inpainting task. During inference, we also utilize image similarity search over image embeddings for prompt selection and novel class filtering to reduce false positive predictions. Ba

Stats

• citations8
• S2 citations—
• github stars0
• HF likes0
• heat score7.16
• arxiv keyimmanuel2024learnable

Related papers

• Retrieval-enhanced Visual Prompt Learning For Few-shot Classification (2023)4.52
• Less Than Few: Self-shot Video Instance Segmentation (2022)2.26
• Towards In-context Scene Understanding (2023)0.00
• Stochastic-based Patch Filtering For Few-shot Learning (2025)0.00
• Fine-grained Retrieval Prompt Tuning (2022)10.07
• Locus: Learning Multiscale 3d-consistent Features From Posed Images (2023)0.00
• What Makes Good Examples For Visual In-context Learning? (2023)3.58
• Elevating All Zero-shot Sketch-based Image Retrieval Through Multimodal Prompt Learning (2024)6.34