Weakly Supervised Semantic Segmentation using Out-of-Distribution Data
- URL: http://arxiv.org/abs/2203.03860v1
- Date: Tue, 8 Mar 2022 05:33:35 GMT
- Title: Weakly Supervised Semantic Segmentation using Out-of-Distribution Data
- Authors: Jungbeom Lee, Seong Joon Oh, Sangdoo Yun, Junsuk Choe, Eunji Kim,
Sungroh Yoon
- Abstract summary: Weakly supervised semantic segmentation (WSSS) methods are often built on pixel-level localization maps.
We propose a novel source of information to distinguish foreground from the background: Out-of-Distribution (OoD) data.
- Score: 50.45689349004041
- License: http://creativecommons.org/licenses/by-nc-sa/4.0/
- Abstract: Weakly supervised semantic segmentation (WSSS) methods are often built on
pixel-level localization maps obtained from a classifier. However, training on
class labels only, classifiers suffer from the spurious correlation between
foreground and background cues (e.g. train and rail), fundamentally bounding
the performance of WSSS. There have been previous endeavors to address this
issue with additional supervision. We propose a novel source of information to
distinguish foreground from the background: Out-of-Distribution (OoD) data, or
images devoid of foreground object classes. In particular, we utilize the hard
OoDs that the classifier is likely to make false-positive predictions. These
samples typically carry key visual features on the background (e.g. rail) that
the classifiers often confuse as foreground (e.g. train), so these cues let
classifiers correctly suppress spurious background cues. Acquiring such hard
OoDs does not require an extensive amount of annotation efforts; it only incurs
a few additional image-level labeling costs on top of the original efforts to
collect class labels. We propose a method, W-OoD, for utilizing the hard OoDs.
W-OoD achieves state-of-the-art performance on Pascal VOC 2012.
Related papers
- Learning to Detour: Shortcut Mitigating Augmentation for Weakly Supervised Semantic Segmentation [7.5856806269316825]
Weakly supervised semantic segmentation (WSSS) employing weak forms of labels has been actively studied to alleviate the annotation cost of acquiring pixel-level labels.
We propose shortcut mitigating augmentation (SMA) for WSSS, which generates synthetic representations of object-background combinations not seen in the training data to reduce the use of shortcut features.
arXiv Detail & Related papers (2024-05-28T13:07:35Z) - Scribble Hides Class: Promoting Scribble-Based Weakly-Supervised
Semantic Segmentation with Its Class Label [16.745019028033518]
We propose a class-driven scribble promotion network, which utilizes both scribble annotations and pseudo-labels informed by image-level classes and global semantics for supervision.
Experiments on the ScribbleSup dataset with different qualities of scribble annotations outperform all the previous methods, demonstrating the superiority and robustness of our method.
arXiv Detail & Related papers (2024-02-27T14:51:56Z) - Virtual Category Learning: A Semi-Supervised Learning Method for Dense
Prediction with Extremely Limited Labels [63.16824565919966]
This paper proposes to use confusing samples proactively without label correction.
A Virtual Category (VC) is assigned to each confusing sample in such a way that it can safely contribute to the model optimisation.
Our intriguing findings highlight the usage of VC learning in dense vision tasks.
arXiv Detail & Related papers (2023-12-02T16:23:52Z) - Towards Single Stage Weakly Supervised Semantic Segmentation [2.28438857884398]
We present a single-stage approach to weakly supervised semantic segmentation.
We use point annotations to generate reliable, on-the-fly pseudo-masks.
We significantly outperform other SOTA WSSS methods on recent real-world datasets.
arXiv Detail & Related papers (2021-06-18T18:34:50Z) - A Closer Look at Self-training for Zero-Label Semantic Segmentation [53.4488444382874]
Being able to segment unseen classes not observed during training is an important technical challenge in deep learning.
Prior zero-label semantic segmentation works approach this task by learning visual-semantic embeddings or generative models.
We propose a consistency regularizer to filter out noisy pseudo-labels by taking the intersections of the pseudo-labels generated from different augmentations of the same image.
arXiv Detail & Related papers (2021-04-21T14:34:33Z) - Background-Aware Pooling and Noise-Aware Loss for Weakly-Supervised
Semantic Segmentation [27.50216933606052]
We address the problem of weakly-supervised semantic segmentation using bounding box annotations.
Background regions are perceptually consistent in part within an image, and this can be leveraged to discriminate foreground and background regions inside object bounding boxes.
We introduce a noise-aware loss (NAL) that makes the networks less susceptible to incorrect labels.
arXiv Detail & Related papers (2021-04-02T06:38:41Z) - Revisiting Deep Local Descriptor for Improved Few-Shot Classification [56.74552164206737]
We show how one can improve the quality of embeddings by leveraging textbfDense textbfClassification and textbfAttentive textbfPooling.
We suggest to pool feature maps by applying attentive pooling instead of the widely used global average pooling (GAP) to prepare embeddings for few-shot classification.
arXiv Detail & Related papers (2021-03-30T00:48:28Z) - Grafit: Learning fine-grained image representations with coarse labels [114.17782143848315]
This paper tackles the problem of learning a finer representation than the one provided by training labels.
By jointly leveraging the coarse labels and the underlying fine-grained latent space, it significantly improves the accuracy of category-level retrieval methods.
arXiv Detail & Related papers (2020-11-25T19:06:26Z) - GPU-based Self-Organizing Maps for Post-Labeled Few-Shot Unsupervised
Learning [2.922007656878633]
Few-shot classification is a challenge in machine learning where the goal is to train a classifier using a very limited number of labeled examples.
We consider the problem of post-labeled few-shot unsupervised learning, a classification task where representations are learned in an unsupervised fashion, to be later labeled using very few annotated examples.
arXiv Detail & Related papers (2020-09-04T13:22:28Z) - Self-Supervised Tuning for Few-Shot Segmentation [82.32143982269892]
Few-shot segmentation aims at assigning a category label to each image pixel with few annotated samples.
Existing meta-learning method tends to fail in generating category-specifically discriminative descriptor when the visual features extracted from support images are marginalized in embedding space.
This paper presents an adaptive framework tuning, in which the distribution of latent features across different episodes is dynamically adjusted based on a self-segmentation scheme.
arXiv Detail & Related papers (2020-04-12T03:53:53Z)
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.