The Devil is in Low-Level Features for Cross-Domain Few-Shot Segmentation

• URL: http://arxiv.org/abs/2503.21150v1
• Date: Thu, 27 Mar 2025 04:37:52 GMT
• Title: The Devil is in Low-Level Features for Cross-Domain Few-Shot Segmentation
• Authors: Yuhan Liu, Yixiong Zou, Yuhua Li, Ruixuan Li,
• Abstract summary: Cross-Domain Few-Shot (CDFSS) is proposed to transfer the pixel-level segmentation capabilities learned from large-scale source-domain datasets to downstream target-domain datasets.<n>We focus on a well-observed but unresolved phenomenon in CDFSS: for target domains, segmentation performance peaks at the very early epochs, and declines sharply as the source-domain training proceeds.<n>We propose a method that includes two plug-and-play modules: one to flatten the loss landscapes for low-level features during source-domain training as a novel sharpness-aware method, and the other to directly supplement target-
• Score: 22.443834719018795
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Cross-Domain Few-Shot Segmentation (CDFSS) is proposed to transfer the pixel-level segmentation capabilities learned from large-scale source-domain datasets to downstream target-domain datasets, with only a few annotated images per class. In this paper, we focus on a well-observed but unresolved phenomenon in CDFSS: for target domains, particularly those distant from the source domain, segmentation performance peaks at the very early epochs, and declines sharply as the source-domain training proceeds. We delve into this phenomenon for an interpretation: low-level features are vulnerable to domain shifts, leading to sharper loss landscapes during the source-domain training, which is the devil of CDFSS. Based on this phenomenon and interpretation, we further propose a method that includes two plug-and-play modules: one to flatten the loss landscapes for low-level features during source-domain training as a novel sharpness-aware minimization method, and the other to directly supplement target-domain information to the model during target-domain testing by low-level-based calibration. Extensive experiments on four target datasets validate our rationale and demonstrate that our method surpasses the state-of-the-art method in CDFSS signifcantly by 3.71% and 5.34% average MIoU in 1-shot and 5-shot scenarios, respectively.

Related papers

• Lightweight Frequency Masker for Cross-Domain Few-Shot Semantic Segmentation [9.365590675168589]
  Cross-domain few-shot segmentation (CD-FSS) is proposed to first pre-train the model on a large-scale source-domain dataset. We propose a lightweight frequency masker, which further reduces channel correlations by an Amplitude-Phase Masker (APM) module and an Adaptive Channel Phase Attention (ACPA) module.
  arXiv  Detail & Related papers  (2024-10-29T15:31:27Z)
• Flatten Long-Range Loss Landscapes for Cross-Domain Few-Shot Learning [8.729669566501103]
  Cross-domain few-shot learning aims to acquire knowledge from limited training data in the target domain. We introduce a new normalization layer that replaces the original one in both CNNs and ViTs. Our method achieves performance improvements of up to 9% compared to the current best approaches on individual datasets.
  arXiv  Detail & Related papers  (2024-03-01T14:44:41Z)
• Cross-Domain Few-Shot Segmentation via Iterative Support-Query Correspondence Mining [81.09446228688559]
  Cross-Domain Few-Shots (CD-FSS) poses the challenge of segmenting novel categories from a distinct domain using only limited exemplars. We propose a novel cross-domain fine-tuning strategy that addresses the challenging CD-FSS tasks.
  arXiv  Detail & Related papers  (2024-01-16T14:45:41Z)
• Cross-Level Distillation and Feature Denoising for Cross-Domain Few-Shot Classification [49.36348058247138]
  We tackle the problem of cross-domain few-shot classification by making a small proportion of unlabeled images in the target domain accessible in the training stage. We meticulously design a cross-level knowledge distillation method, which can strengthen the ability of the model to extract more discriminative features in the target dataset. Our approach can surpass the previous state-of-the-art method, Dynamic-Distillation, by 5.44% on 1-shot and 1.37% on 5-shot classification tasks.
  arXiv  Detail & Related papers  (2023-11-04T12:28:04Z)
• Self-training through Classifier Disagreement for Cross-Domain Opinion Target Extraction [62.41511766918932]
  Opinion target extraction (OTE) or aspect extraction (AE) is a fundamental task in opinion mining. Recent work focus on cross-domain OTE, which is typically encountered in real-world scenarios. We propose a new SSL approach that opts for selecting target samples whose model output from a domain-specific teacher and student network disagrees on the unlabelled target data.
  arXiv  Detail & Related papers  (2023-02-28T16:31:17Z)
• High-level semantic feature matters few-shot unsupervised domain adaptation [15.12545632709954]
  We propose a novel task-specific semantic feature learning method (TSECS) for FS-UDA. TSECS learns high-level semantic features for image-to-class similarity measurement. We show that the proposed method significantly outperforms SOTA methods in FS-UDA by a large margin.
  arXiv  Detail & Related papers  (2023-01-05T08:39:52Z)
• I2F: A Unified Image-to-Feature Approach for Domain Adaptive Semantic Segmentation [55.633859439375044]
  Unsupervised domain adaptation (UDA) for semantic segmentation is a promising task freeing people from heavy annotation work. Key idea to tackle this problem is to perform both image-level and feature-level adaptation jointly. This paper proposes a novel UDA pipeline for semantic segmentation that unifies image-level and feature-level adaptation.
  arXiv  Detail & Related papers  (2023-01-03T15:19:48Z)
• Stagewise Unsupervised Domain Adaptation with Adversarial Self-Training for Road Segmentation of Remote Sensing Images [93.50240389540252]
  Road segmentation from remote sensing images is a challenging task with wide ranges of application potentials. We propose a novel stagewise domain adaptation model called RoadDA to address the domain shift (DS) issue in this field. Experiment results on two benchmarks demonstrate that RoadDA can efficiently reduce the domain gap and outperforms state-of-the-art methods.
  arXiv  Detail & Related papers  (2021-08-28T09:29:14Z)
• Physically-Constrained Transfer Learning through Shared Abundance Space for Hyperspectral Image Classification [14.840925517957258]
  We propose a new transfer learning scheme to bridge the gap between the source and target domains. The proposed method is referred to as physically-constrained transfer learning through shared abundance space.
  arXiv  Detail & Related papers  (2020-08-19T17:41:37Z)
• Alleviating Semantic-level Shift: A Semi-supervised Domain Adaptation Method for Semantic Segmentation [97.8552697905657]
  A key challenge of this task is how to alleviate the data distribution discrepancy between the source and target domains. We propose Alleviating Semantic-level Shift (ASS), which can successfully promote the distribution consistency from both global and local views. We apply our ASS to two domain adaptation tasks, from GTA5 to Cityscapes and from Synthia to Cityscapes.
  arXiv  Detail & Related papers  (2020-04-02T03:25:05Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.