Filters








1,066 Hits in 5.6 sec

Semantically Tied Paired Cycle Consistency for Any-Shot Sketch-based Image Retrieval [article]

Anjan Dutta, Zeynep Akata
2020 arXiv   pre-print
In this paper, we address any-shot, i.e. zero-shot and few-shot, sketch-based image retrieval (SBIR) tasks, where we introduce the few-shot setting for SBIR.  ...  For solving these tasks, we propose a semantically aligned paired cycle-consistent generative adversarial network (SEM-PCYC) for any-shot SBIR, where each branch of the generative adversarial network maps  ...  The TITAN Xp and TITAN V used for this research were donated by the NVIDIA Corporation.  ... 
arXiv:2006.11397v1 fatcat:4tkatm4h5jdzjacdx23x7jbd7y

Semantically Tied Paired Cycle Consistency for Any-Shot Sketch-Based Image Retrieval

Anjan Dutta, Zeynep Akata
2020 International Journal of Computer Vision  
In this paper, we address any-shot, i.e. zero-shot and few-shot, sketch-based image retrieval (SBIR) tasks, where we introduce the few-shot setting for SBIR.  ...  For solving these tasks, we propose a semantically aligned paired cycle-consistent generative adversarial network (SEM-PCYC) for any-shot SBIR, where each branch of the generative adversarial network maps  ...  The TITAN Xp and TITAN V used for this research were donated by the NVIDIA Corporation.  ... 
doi:10.1007/s11263-020-01350-x fatcat:chi7krnz3zhmbbyw7huiosh3ey

Semantically Tied Paired Cycle Consistency for Zero-Shot Sketch-based Image Retrieval [article]

Anjan Dutta, Zeynep Akata
2019 arXiv   pre-print
Zero-shot sketch-based image retrieval (SBIR) is an emerging task in computer vision, allowing to retrieve natural images relevant to sketch queries that might not been seen in the training phase.  ...  In this work, we propose a semantically aligned paired cycle-consistent generative (SEM-PCYC) model for zero-shot SBIR, where each branch maps the visual information to a common semantic space via an adversarial  ...  The Titan Xp and Titan V used for this research were donated by the NVIDIA Corporation.  ... 
arXiv:1903.03372v1 fatcat:ig3bjce4kfeefjnqyhgmujp7dy

Three-Stream Joint Network for Zero-Shot Sketch-Based Image Retrieval [article]

Yu-Wei Zhan, Xin Luo, Yongxin Wang, Zhen-Duo Chen, Xin-Shun Xu
2022 arXiv   pre-print
The Zero-Shot Sketch-based Image Retrieval (ZS-SBIR) is a challenging task because of the large domain gap between sketches and natural images as well as the semantic inconsistency between seen and unseen  ...  And most works reduce domain gap by mapping sketches and natural images into a common high-level space using constructed sketch-image pairs, which ignore the unpaired information between images and sketches  ...  SEM-PCYC [8] proposes a semantically tied paired cycle consistency generation model that maps the visual information of sketches and images to a common semantic space by adversarial training.  ... 
arXiv:2204.05666v1 fatcat:ug5w5236e5bs3ase3e6oyu2syu

StyleMeUp: Towards Style-Agnostic Sketch-Based Image Retrieval [article]

Aneeshan Sain, Ayan Kumar Bhunia, Yongxin Yang, Tao Xiang, Yi-Zhe Song
2021 arXiv   pre-print
Sketch-based image retrieval (SBIR) is a cross-modal matching problem which is typically solved by learning a joint embedding space where the semantic content shared between photo and sketch modalities  ...  With this meta-learning framework, our model can not only disentangle the cross-modal shared semantic content for SBIR, but can adapt the disentanglement to any unseen user style as well, making the SBIR  ...  (ii) We hypothesise that true efficiency for our sketch representation can be judged in a sketch based sketch retrieval problem.  ... 
arXiv:2103.15706v2 fatcat:ukbeu2bpujb53j3pzd5zbdldai

Deep Learning for Free-Hand Sketch: A Survey [article]

Peng Xu, Timothy M. Hospedales, Qiyue Yin, Yi-Zhe Song, Tao Xiang, Liang Wang
2022 arXiv   pre-print
(iii) Promotion of future work via a discussion of bottlenecks, open problems, and potential research directions for the community.  ...  The recent prevalence of touchscreen devices has made sketch creation a much easier task than ever and consequently made sketch-oriented applications increasingly popular.  ...  In recent years, motivated by the zero-shot setting for supervised photo retrieval [241] , zero-shot sketch-based image retrieval (ZS-SBIR) has also been studied [65] , [216] , [242] - [255] .  ... 
arXiv:2001.02600v3 fatcat:lek5sivzsrat3i52lqh2eifnia

Adversarial Open Domain Adaptation for Sketch-to-Photo Synthesis [article]

Xiaoyu Xiang, Ding Liu, Xiao Yang, Yiheng Zhu, Xiaohui Shen, Jan P. Allebach
2021 arXiv   pre-print
Compared with the recent competing methods, our approach shows impressive results in synthesizing realistic color, texture, and maintaining the geometric composition for various categories of open-domain  ...  sketches.  ...  Semantically tied paired cycle consistency for zero-shot sketch-based image retrieval.  ... 
arXiv:2104.05703v2 fatcat:m4kkb6kje5botkae4xe25pxdu4

Learning by Aligning: Visible-Infrared Person Re-identification using Cross-Modal Correspondences [article]

Hyunjong Park, Sanghoon Lee, Junghyup Lee, Bumsub Ham
2021 arXiv   pre-print
We address the problem of visible-infrared person re-identification (VI-reID), that is, retrieving a set of person images, captured by visible or infrared cameras, in a cross-modal setting.  ...  This also encourages pixel-wise associations between cross-modal local features, further facilitating discriminative feature learning for VI-reID.  ...  For example, they synthesize novel IR person images, with an identity-preserving constraint [34] or cycle consistency [35] , given RGB inputs, in order to compare person images with the same modality  ... 
arXiv:2108.07422v1 fatcat:6vu7sfhfe5ah7a3jo6pxznl55q

Search in audiovisual broadcast archives

Bouke Huurnink
2011 SIGIR Forum  
For example, a news editor might wish to reuse footage from overseas services for the evening news, or a documentary maker describing the history of the Christmas tradition might desire shots of Christmas  ...  In Part II we study how the results of their searches may be improved through applying state-ofthe-art methods in video retrieval.  ...  [147] provide a categorization specific to content-based image retrieval queries: target (or known-item) search (when the user has a specific image in mind), category search (retrieving an arbitrary  ... 
doi:10.1145/1988852.1988871 fatcat:ckotehsrpnbjpcyf63rvnyui3i

The relevance of redundancy in multimodal documents

Olli Philippe Lautenbacher
2019 Linguistica Antverpiensia, New Series – Themes in Translation Studies  
What is advocated here is the idea of a recursive reading process consisting of three phases (perception, construction and integration) and that this process is based on the detection of a salient series  ...  In the translation process, any change within this redundancy system, such as a modification in the balance between endophora and exophora, might alter the overall reception experience.  ...  The actual integration following each processing cycle consists of "freezing" the most highly activated nodes of the net.  ... 
doi:10.52034/lanstts.v17i0.462 fatcat:lcer2tfuzjgqxgrg2fhvr6clou

Learning Transferable Visual Models From Natural Language Supervision [article]

Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, Ilya Sutskever
2021 arXiv   pre-print
million (image, text) pairs collected from the internet.  ...  For instance, we match the accuracy of the original ResNet-50 on ImageNet zero-shot without needing to use any of the 1.28 million training examples it was trained on.  ...  Over 20 years ago Mori et al. (1999) explored improving content based image retrieval by training a model to predict the nouns and adjectives in text documents paired with images.  ... 
arXiv:2103.00020v1 fatcat:sbmk67xkqnfsxldwsfkbxzqdui

Proceedings of eNTERFACE 2015 Workshop on Intelligent Interfaces [article]

Matei Mancas, Christian Frisson, Joëlle Tilmanne, Nicolas d'Alessandro, Petr Barborka, Furkan Bayansar, Francisco Bernard, Rebecca Fiebrink, Alexis Heloir, Edgar Hemery, Sohaib Laraba, Alexis Moinet (+58 others)
2018 arXiv   pre-print
The team would like to thank Metapraxis for supporting this project and lending us one of the tablets for the experiments.  ...  The team would also like to thank the DeVisu laboratory for lending us the Tobii eyetracking glasses. The team would like to thank the TCTS laboratory for the WiFi hotspots.  ...  In an analogous manner, one can hence think about "sketching" a representation of the soundtrack of the video shot to be retrieved.  ... 
arXiv:1801.06349v1 fatcat:qauytivdq5axxis2xlknp3r2ne

WhittleSearch: Interactive Image Search with Relative Attribute Feedback

Adriana Kovashka, Devi Parikh, Kristen Grauman
2015 International Journal of Computer Vision  
For example, perusing image results for a query "black shoes", the user might state, "Show me shoe images like these, but sportier."  ...  We propose a novel mode of feedback for image search, where a user describes which properties of exemplar images should be adjusted in order to more closely match his/her mental model of the image sought  ...  Acknowledgements We thank the anonymous reviewers for their helpful feedback and suggestions.  ... 
doi:10.1007/s11263-015-0814-0 fatcat:ekkzyhbkejavfkvfwkezgu4u2m

What is episodic memory if it is a natural kind?

Sen Cheng, Markus Werning
2015 Synthese  
We propose to conceive of episodic memory as a knowledge-like state that is identified with an experientially based mnemonic representation of an episode that allows for a mnemonic simulation thereof.  ...  The argumentation proceeds along three cornerstones: First, psychological evidence suggests that a violation of any of the proposed conditions for episodic memory amounts to a deficiency of episodic memory  ...  Acknowledgements: We thank Thomas Suddendorf for helpful discussions and Kevin Reuter for comments on the manuscript.  ... 
doi:10.1007/s11229-014-0628-6 fatcat:rawyyi3i2bcaxcyiua4oots6pq

Towards holistic scene understanding: Semantic segmentation and beyond [article]

Panagiotis Meletis
2022 arXiv   pre-print
Chapter 3 focuses on enriching semantic segmentation with weak supervision and proposes a weakly-supervised algorithm for training with bounding box-level and image-level supervision instead of only with  ...  This framework achieves consistent increases in performance metrics and semantic knowledgeability by exploiting various scene understanding datasets.  ...  A PSV match was held the same day and a big party was organized for the win.  ... 
arXiv:2201.07734v1 fatcat:qdqnjqn75rff7kyja2iwer75my
« Previous Showing results 1 — 15 out of 1,066 results