AMP: Adaptive Masked Proxies for Few-Shot Segmentation

Siam, Mennatullah; Oreshkin, Boris N.; Jägersand, Martin

doi:10.1109/iccv.2019.00535

Cited by 181 publications

(135 citation statements)

References 30 publications

Supporting

Mentioning

131

Contrasting

Order By: Relevance

“…All of which extend such dual branch structure to achieve a substantial performance improvement. In the AMP model, Siam et al [21] replaces the guidance branch with a multiresolution weight. Moreover, SG-One [22] proposed a Masked Average Pooling block (MAP) to extract the representative vectors of support objects.…”

Section: Related Workmentioning

confidence: 99%

Incorporating Depth Information into Few-Shot Semantic Segmentation

Zhang¹,

Sidibé

Morel³

et al. 2021

2020 25th International Conference on Pattern Recognition (ICPR)

View full text Add to dashboard Cite

Few-shot segmentation presents a significant challenge for semantic scene understanding under limited supervision. Namely, this task targets at generalizing the segmentation ability of the model to new categories given a few samples. In order to obtain complete scene information, we extend the RGB-centric methods to take advantage of complementary depth information. In this paper, we propose a two-stream deep neural network based on metric learning. Our method, known as RDNet, learns class-specific prototype representations within RGB and depth embedding spaces, respectively. The learned prototypes provide effective semantic guidance on the corresponding RGB and depth query image, leading to more accurate performance. Moreover, we build a novel outdoor scene dataset, known as Cityscapes-3 i , using labeled RGB images and depth images from the Cityscapes dataset. We also perform ablation studies to explore the effective use of depth information in few-shot segmentation tasks. Experiments on Cityscapes-3 i show that our method achieves excellent results with visual and complementary geometric cues from only a few labeled examples.

show abstract

Section: Related Workmentioning

confidence: 99%

Incorporating Depth Information into Few-Shot Semantic Segmentation

Zhang¹,

Sidibé

Morel³

et al. 2021

2020 25th International Conference on Pattern Recognition (ICPR)

View full text Add to dashboard Cite

show abstract

“…Few-shot learning is a related area where, after learning on a set of base classes with abundant examples, new tasks are given with only few examples of novel (unseen) classes [54,45,14]. In fact, few-shot learning has been applied to both object detection [21,9] and semantic segmentation [32,59,28,42]. However, we are also using a large training set of unlabeled images, while our target classes may not be unseen.…”

Section: Related Workmentioning

confidence: 99%

Training object detectors from few weakly-labeled and many unlabeled images

Yang

Shi

et al. 2021

Pattern Recognition

View full text Add to dashboard Cite

Weakly-supervised object detection attempts to limit the amount of supervision by dispensing the need for bounding boxes, but still assumes image-level labels on the entire training set are available. In this work, we study the problem of training an object detector from one or few clean images with image-level labels and a larger set of completely unlabeled images. This is an extreme case of semisupervised learning where the labeled data are not enough to bootstrap the learning of a classifier or detector.Our solution is to use a standard weakly-supervised pipeline to train a student model from image-level pseudolabels generated on the unlabeled set by a teacher model, bootstrapped by region-level similarities to clean labeled images. By using the recent pipeline of PCL [47] and more unlabeled images, we achieve performance competitive or superior to many state of the art weakly-supervised detection solutions.

show abstract

“…Few-shot object segmentation [15,29,35] has received much attention recently due to its advantages in learning novel categories [8,9] without much annotations. Most previous approaches [30,42] follow the metric-based few-shot learning scheme and make great efforts on developing robust feature embedding to measure the pixel-wise similarity between the object from the support image and the query one. However, current few-shot segmentation [27,38,43] only considers a simple case, i.e., segmenting one or two objects from unseen categories in the given query image, which usually does not work well for the real scenario where pixels from dozens of unseen categories appear.…”

Section: Related Workmentioning

confidence: 99%

Meta Parsing Networks: Towards Generalized Few-shot Scene Parsing with Adaptive Metric Learning

Wei

Yang

2020

Proceedings of the 28th ACM International Conference on Multimedia

View full text Add to dashboard Cite

Recent progress in few-shot segmentation usually aims at performing novel object segmentation using a few annotated examples as guidance. In this work, we advance this few-shot segmentation paradigm towards a more challenging yet general scenario, i.e., Generalized Few-shot Scene Parsing (GFSP). In this task, we take a fully annotated image as guidance to segment all pixels in a query image. Our mission is to study a generalizable and robust segmentation network from the meta-learning perspective so that both seen and unseen categories can be correctly recognized. Different from previous practices, this task performs segmentation on a joint label space consisting of both previously seen and novel categories. Moreover, pixels from these multiple categories need to be simultaneously taken into account, which is actually not well explored before. Accordingly, we present Meta Parsing Networks (MPNet) to better exploit the guidance information in the support set. Our MPNet contains two basic modules, i.e., the Adaptive Deep Metric Learning (ADML) module and the Contrastive Inter-class Distraction (CID) module. Specially, the ADML takes the annotated pixels from the support image as the guidance and adaptively produces high-quality prototypes for learning a deep comparison metric. In addition, MPNet further introduces the CID module learning to enlarge the feature discrepancy of different categories in the embedding space, leading the MPNet to generate more discriminative feature embeddings. We conduct experiments on two newly constructed benchmarks, i.e., GFSP-Cityscapes and GFSP-Pascal-Context. Extensive ablation studies well demonstrate the effectiveness and generalization ability of our MPNet. CCS CONCEPTS • Computing methodologies → Computer vision.

show abstract

AMP: Adaptive Masked Proxies for Few-Shot Segmentation

Cited by 181 publications

References 30 publications

Incorporating Depth Information into Few-Shot Semantic Segmentation

Incorporating Depth Information into Few-Shot Semantic Segmentation

Training object detectors from few weakly-labeled and many unlabeled images

Meta Parsing Networks: Towards Generalized Few-shot Scene Parsing with Adaptive Metric Learning

Contact Info

Product

Resources

About