Knowledge Distillation for Semantic Segmentation: A Label Space Unification Approach

• URL: http://arxiv.org/abs/2502.19177v1
• Date: Wed, 26 Feb 2025 14:33:33 GMT
• Title: Knowledge Distillation for Semantic Segmentation: A Label Space Unification Approach
• Authors: Anton Backhaus, Thorsten Luettel, Mirko Maehlisch,
• Abstract summary: We propose a knowledge distillation approach that also serves as a label space unification method for semantic segmentation.<n>A teacher model is trained on a source dataset with a given taxonomy, then used to pseudo-label additional data for which ground truth labels of a related label space exist.<n>Our ground truth-corrected pseudo-labels span over 12 and 7 public datasets with 388.230 and 18.558 images for the urban and off-road domains, respectively.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: An increasing number of datasets sharing similar domains for semantic segmentation have been published over the past few years. But despite the growing amount of overall data, it is still difficult to train bigger and better models due to inconsistency in taxonomy and/or labeling policies of different datasets. To this end, we propose a knowledge distillation approach that also serves as a label space unification method for semantic segmentation. In short, a teacher model is trained on a source dataset with a given taxonomy, then used to pseudo-label additional data for which ground truth labels of a related label space exist. By mapping the related taxonomies to the source taxonomy, we create constraints within which the model can predict pseudo-labels. Using the improved pseudo-labels we train student models that consistently outperform their teachers in two challenging domains, namely urban and off-road driving. Our ground truth-corrected pseudo-labels span over 12 and 7 public datasets with 388.230 and 18.558 images for the urban and off-road domains, respectively, creating the largest compound datasets for autonomous driving to date.

Related papers

• COARSE: Collaborative Pseudo-Labeling with Coarse Real Labels for Off-Road Semantic Segmentation [49.267650162344765]
  COARSE is a semi-supervised domain adaptation framework for off-road semantic segmentation. We bridge domain gaps with complementary pixel-level and patch-level decoders, enhanced by a collaborative pseudo-labeling strategy.
  arXiv  Detail & Related papers  (2025-03-05T22:25:54Z)
• Scribbles for All: Benchmarking Scribble Supervised Segmentation Across Datasets [51.74296438621836]
  We introduce Scribbles for All, a label and training data generation algorithm for semantic segmentation trained on scribble labels. The main limitation of scribbles as source for weak supervision is the lack of challenging datasets for scribble segmentation. Scribbles for All provides scribble labels for several popular segmentation datasets and provides an algorithm to automatically generate scribble labels for any dataset with dense annotations.
  arXiv  Detail & Related papers  (2024-08-22T15:29:08Z)
• You can't handle the (dirty) truth: Data-centric insights improve pseudo-labeling [60.27812493442062]
  We show the importance of investigating labeled data quality to improve any pseudo-labeling method. Specifically, we introduce a novel data characterization and selection framework called DIPS to extend pseudo-labeling. We demonstrate the applicability and impact of DIPS for various pseudo-labeling methods across an extensive range of real-world datasets.
  arXiv  Detail & Related papers  (2024-06-19T17:58:40Z)
• From Isolated Islands to Pangea: Unifying Semantic Space for Human Action Understanding [50.412121156940294]
  Action understanding can be formed as the mapping from the physical space to the semantic space. We propose a novel model mapping from the physical space to semantic space to fully use Pangea.
  arXiv  Detail & Related papers  (2023-04-02T15:04:43Z)
• Label Name is Mantra: Unifying Point Cloud Segmentation across Heterogeneous Datasets [17.503843467554592]
  We propose a principled approach that supports learning from heterogeneous datasets with different label sets. Our idea is to utilize a pre-trained language model to embed discrete labels to a continuous latent space with the help of their label names. Our model outperforms the state-of-the-art by a large margin.
  arXiv  Detail & Related papers  (2023-03-19T06:14:22Z)
• Revisiting Self-Training with Regularized Pseudo-Labeling for Tabular Data [0.0]
  We revisit self-training which can be applied to any kind of algorithm including gradient boosting decision tree. We propose a novel pseudo-labeling approach that regularizes the confidence scores based on the likelihoods of the pseudo-labels.
  arXiv  Detail & Related papers  (2023-02-27T18:12:56Z)
• Learning Semantic Segmentation from Multiple Datasets with Label Shifts [101.24334184653355]
  This paper proposes UniSeg, an effective approach to automatically train models across multiple datasets with differing label spaces. Specifically, we propose two losses that account for conflicting and co-occurring labels to achieve better generalization performance in unseen domains.
  arXiv  Detail & Related papers  (2022-02-28T18:55:19Z)
• Joining datasets via data augmentation in the label space for neural networks [6.036150783745836]
  We propose a new technique leveraging artificially created knowledge graph, recurrent neural networks and policy gradient that successfully achieve the dataset joining in the label space. Empirical results on both image and text classification justify the validity of our approach.
  arXiv  Detail & Related papers  (2021-06-17T06:08:11Z)
• Graphonomy: Universal Image Parsing via Graph Reasoning and Transfer [140.72439827136085]
  We propose a graph reasoning and transfer learning framework named "Graphonomy" It incorporates human knowledge and label taxonomy into the intermediate graph representation learning beyond local convolutions. It learns the global and structured semantic coherency in multiple domains via semantic-aware graph reasoning and transfer.
  arXiv  Detail & Related papers  (2021-01-26T08:19:03Z)

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.