Filters








10,673 Hits in 3.8 sec

Zero-Shot Semantic Segmentation [article]

Maxime Bucher, Tuan-Hung Vu, Matthieu Cord, Patrick Pérez
2019 arXiv   pre-print
In this paper, we introduce the new task of zero-shot semantic segmentation: learning pixel-wise classifiers for never-seen object categories with zero training examples.  ...  On the two standard segmentation datasets, Pascal-VOC and Pascal-Context, we propose zero-shot benchmarks and set competitive baselines.  ...  Conclusion In this work, we introduced a deep model to deal with the task of zero-shot semantic segmentation.  ... 
arXiv:1906.00817v2 fatcat:st5vlsaxmzfahbei6uivmt3sma

Decoupling Zero-Shot Semantic Segmentation [article]

Jian Ding, Nan Xue, Gui-Song Xia, Dengxin Dai
2022 arXiv   pre-print
Zero-shot semantic segmentation (ZS3) aims to segment the novel categories that have not been seen in the training.  ...  . 2) a zero-shot classification task on segments.  ...  Following the fully supervised semantic segmentation models [10, 11, 36] and zero-shot classification models [1, 24, 30, 47, 60] , these works formulate zero-shot semantic segmentation as a pixel-level  ... 
arXiv:2112.07910v2 fatcat:2mlfmsvrszbi3awgqy6r6igpo4

Generative Zero-Shot Learning for Semantic Segmentation of 3D Point Cloud [article]

Björn Michele, Alexandre Boulch, Gilles Puy, Maxime Bucher, Renaud Marlet
2021 arXiv   pre-print
While there has been a number of studies on Zero-Shot Learning (ZSL) for 2D images, its application to 3D data is still recent and scarce, with just a few methods limited to classification.  ...  For semantic segmentation, we created three benchmarks for evaluating this new ZSL task, using S3DIS, ScanNet and SemanticKITTI.  ...  Zero-shot point cloud segmentation on SemanticKITTI.  ... 
arXiv:2108.06230v4 fatcat:drda76vitfaupc43ozk5ar7sva

From Pixel to Patch: Synthesize Context-aware Features for Zero-shot Semantic Segmentation [article]

Zhangxuan Gu, Siyuan Zhou, Li Niu, Zihan Zhao, Liqing Zhang
2022 arXiv   pre-print
Code is available at https://github.com/bcmi/CaGNetv2-Zero-Shot-Semantic-Segmentation.  ...  Thus, we focus on zero-shot semantic segmentation, which aims to segment unseen objects with only category-level semantic representations provided for unseen categories.  ...  So it is in high demand to design effective zero-shot semantic segmentation methods. The setting of zero-shot semantic segmentation [23] is similar to that of zero-shot classification.  ... 
arXiv:2009.12232v4 fatcat:gcjhzj42cbaklh5ebq3pzszn6q

Exploiting a Joint Embedding Space for Generalized Zero-Shot Semantic Segmentation [article]

Donghyeon Baek, Youngmin Oh, Bumsub Ham
2021 arXiv   pre-print
We address the problem of generalized zero-shot semantic segmentation (GZS3) predicting pixel-wise semantic labels for seen and unseen classes.  ...  To this end, we leverage visual and semantic encoders to learn a joint embedding space, where the semantic encoder transforms semantic features to semantic prototypes that act as centers for visual features  ...  Zero-shot semantic segmentation. Recently, there are many attempts to extend ZSL methods for image classification to the task of semantic segmentation.  ... 
arXiv:2108.06536v1 fatcat:japzmaebdfhrrbgooky6p3k4xy

Learning unbiased zero-shot semantic segmentation networks via transductive transfer [article]

Haiyang Liu, Yichen Wang, Jiayi Zhao, Guowu Yang, Fengmao Lv
2020 arXiv   pre-print
In this paper, we propose an easy-to-implement transductive approach to alleviate the prediction bias in zero-shot semantic segmentation.  ...  Since it is impractical to collect labeled data for all categories, how to conduct zero-shot learning in semantic segmentation establishes an important problem.  ...  For zero-shot semantic segmentation, the previous methods propose to incorporate semantic embeddings of categories to semantic segmentation neural networks [7] .  ... 
arXiv:2007.00515v1 fatcat:qxgahnv6urgcpizsdcm6zc5xii

SIGN: Spatial-information Incorporated Generative Network for Generalized Zero-shot Semantic Segmentation [article]

Jiaxin Cheng, Soumyaroop Nandi, Prem Natarajan, Wael Abd-Almageed
2021 arXiv   pre-print
Unlike conventional zero-shot classification, zero-shot semantic segmentation predicts a class label at the pixel level instead of the image level.  ...  When solving zero-shot semantic segmentation problems, the need for pixel-level prediction with surrounding context motivates us to incorporate spatial information using positional encoding.  ...  Zero-shot Semantic Segmentation Bucher et al.  ... 
arXiv:2108.12517v1 fatcat:ui7no7xnxvhuxfzdrv7l7iq2w4

Language-Level Semantics Conditioned 3D Point Cloud Segmentation [article]

Bo Liu, Shuang Deng, Qiulei Dong, Zhanyi Hu
2022 arXiv   pre-print
Two benchmarks are also introduced for a newly introduced zero-shot 3D segmentation task, and the results also validate the proposed framework.  ...  In this work, a language-level Semantics Conditioned framework for 3D Point cloud segmentation, called SeCondPoint, is proposed, where language-level semantics are introduced to condition the modeling  ...  For each data split, two different zero-shot setting are introduced, i.e. the conventional zero-shot segmentation and the generalized zero-shot segmentation.  ... 
arXiv:2107.00430v3 fatcat:p536hu7levbe3o63dj5adku2me

Context-aware Feature Generation for Zero-shot Semantic Segmentation [article]

Zhangxuan Gu, and Siyuan Zhou, and Li Niu, and Zihan Zhao, Liqing Zhang
2020 pre-print
Our method achieves state-of-the-art results on three benchmark datasets for zero-shot segmentation. Codes are available at: https://github.com/bcmi/CaGNet-Zero-Shot-Semantic-Segmentation.  ...  To reduce the annotation pressure, we focus on a challenging task named zero-shot semantic segmentation, which aims to segment unseen objects with zero annotations.  ...  Zero-shot Semantic Segmentation: The term zero-shot semantic segmentation appeared in prior works [3, 17, 43, 49] , in which only SPNet [43] and ZS3Net [3] focused on multi-category semantic segmentation  ... 
doi:10.1145/3394171.3413593 arXiv:2008.06893v1 fatcat:yn4dewhoi5hrdjci4bqimj6rsu

Weak Novel Categories without Tears: A Survey on Weak-Shot Learning [article]

Li Niu
2021 arXiv   pre-print
Among them, zero-shot (resp., few-shot) learning explores using zero (resp., a few) training samples for novel categories, which lowers the quantity requirement for novel categories.  ...  ., noisy labels for image classification, image labels for object detection, bounding boxes for segmentation), similar to the definitions in weakly supervised learning.  ...  Recently, zero-shot object detection [73, 10] , zero-shot semantic segmentation [5, 19, 60] , and zero-shot instance segmentation [67] have also been explored.  ... 
arXiv:2110.02651v2 fatcat:hvvniqgukrdktmu4vjj6fefuqq

Language-driven Semantic Segmentation [article]

Boyi Li and Kilian Q. Weinberger and Serge Belongie and Vladlen Koltun and René Ranftl
2022 arXiv   pre-print
We demonstrate that our approach achieves highly competitive zero-shot performance compared to existing zero- and few-shot semantic segmentation methods, and even matches the accuracy of traditional segmentation  ...  We present LSeg, a novel model for language-driven semantic image segmentation.  ...  This enables flexible synthesis of zero-shot semantic segmentation models on the fly.  ... 
arXiv:2201.03546v2 fatcat:hb5v4ks5zjayxffx2t5xaxcmzy

The devil is in the labels: Semantic segmentation from sentences [article]

Wei Yin, Yifan Liu, Chunhua Shen, Anton van den Hengel, Baichuan Sun
2022 arXiv   pre-print
We propose an approach to semantic segmentation that achieves state-of-the-art supervised performance when applied in a zero-shot setting.  ...  By fine-tuning the model on standard semantic segmentation datasets, we also achieve a significant improvement over the state-of-the-art supervised segmentation on NYUD-V2 and PASCAL-context at 60% and  ...  Applying this new vector label within the already trained model enables zero-shot segmentation of the corresponding class. Figure 1 shows some zero-shot semantic segmentation examples.  ... 
arXiv:2202.02002v1 fatcat:ptgn5ugyevet7jlbics4s4lmp4

A fusion scheme of visual and auditory modalities for event detection in sports video

Min Xu, Ling-Yu Duan, Chang-Sheng Xu, Qi Tian
2003 2003 International Conference on Multimedia and Expo. ICME '03. Proceedings (Cat. No.03TH8698)  
Among major shot classes we perform classification of the different auditory signal segments (i.e. silence, hitting ball, applause, commentator speech) with the goal of detecting events with strong semantic  ...  The proposed scheme is built upon semantic shot classification, where we classify video shots into several major or interesting classes, each of which has clear semantic meanings.  ...  Hence, we make use of zero crossings to distinguish between applause and commentator speech. Figure 4 (b) shows the ZCR of an audio segment from tennis video.  ... 
doi:10.1109/icme.2003.1220922 dblp:conf/icmcs/XuDXT03 fatcat:6gxfz64ayzb7nfg65445zjtwbm

ConQX: Semantic Expansion of Spoken Queries for Intent Detection based on Conditioned Text Generation [article]

Eyup Halit Yilmaz, Cagri Toraman
2021 arXiv   pre-print
We then apply zero-shot, one-shot, and few-shot learning. We lastly use the expanded queries to fine-tune BERT and RoBERTa for intent detection.  ...  The experimental results show that the performance of intent detection can be improved by our semantic expansion method.  ...  The number of generated tokens is approximated to the number of input tokens. • ConQX (zero-shot): Our semantic expansion method with zero-shot learning.  ... 
arXiv:2109.00729v1 fatcat:a5shjralr5clhi6ghoz5zmy5bq

Multi-Layer Cross Loss Model for Zero-Shot Human Activity Recognition [chapter]

Tong Wu, Yiqiang Chen, Yang Gu, Jiwei Wang, Siyu Zhang, Zhanghu Zhechen
2020 Lecture Notes in Computer Science  
Zero-shot learning aims at solving this problem. In this paper, we propose a novel model termed Multi-Layer Cross Loss Model (MLCLM).  ...  Experiments show that our model outperforms other state-of-the-art methods significantly in zero-shot human activity recognition.  ...  To adopt zero-shot learning to HAR, the semantic space is essential for the three datasets.  ... 
doi:10.1007/978-3-030-47426-3_17 fatcat:kr7d7aqokfhvtmgwtcuepsaqm4
« Previous Showing results 1 — 15 out of 10,673 results