Zobrazeno 1 - 10
of 831
pro vyhledávání: '"P. Chum"'
This work proposes a novel method for object co-segmentation, i.e. pixel-level localization of a common object in a set of images, that uses no pixel-level supervision for training. Two pre-trained Vision Transformer (ViT) models are exploited: Image
Externí odkaz:
http://arxiv.org/abs/2410.13582
Single-source domain generalization attempts to learn a model on a source domain and deploy it to unseen target domains. Limiting access only to source domain data imposes two key challenges - how to train a model that can generalize and how to verif
Externí odkaz:
http://arxiv.org/abs/2409.19774
Universal image representations are critical in enabling real-world fine-grained and instance-level recognition applications, where objects and entities from any domain must be identified at large scale. Despite recent advances, existing methods fail
Externí odkaz:
http://arxiv.org/abs/2406.08332
Autor:
Psomas, Bill, Kakogeorgiou, Ioannis, Efthymiadis, Nikos, Tolias, Giorgos, Chum, Ondrej, Avrithis, Yannis, Karantzalos, Konstantinos
This work introduces composed image retrieval to remote sensing. It allows to query a large image archive by image examples alternated by a textual description, enriching the descriptive power over unimodal queries, either visual or textual. Various
Externí odkaz:
http://arxiv.org/abs/2405.15587
Publikováno v:
Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2023, pp. 11153-11163
Image retrieval methods based on CNN descriptors rely on metric learning from a large number of diverse examples of positive and negative image pairs. Domains, such as night-time images, with limited availability and variability of training data suff
Externí odkaz:
http://arxiv.org/abs/2309.16351
Autor:
Ypsilantis, Nikolaos-Antonios, Chen, Kaifeng, Cao, Bingyi, Lipovský, Mário, Dogan-Schönberger, Pelin, Makosa, Grzegorz, Bluntschli, Boris, Seyedhosseini, Mojtaba, Chum, Ondřej, Araujo, André
Fine-grained and instance-level recognition methods are commonly trained and evaluated on specific domains, in a model per domain scenario. Such an approach, however, is impractical in real large-scale applications. In this work, we address the probl
Externí odkaz:
http://arxiv.org/abs/2309.01858
Autor:
Ayyubi, Hammad A., Thomas, Christopher, Chum, Lovish, Lokesh, Rahul, Chen, Long, Niu, Yulei, Lin, Xudong, Feng, Xuande, Koo, Jaywon, Ray, Sounak, Chang, Shih-Fu
Events describe happenings in our world that are of importance. Naturally, understanding events mentioned in multimedia content and how they are related forms an important way of comprehending our world. Existing literature can infer if events across
Externí odkaz:
http://arxiv.org/abs/2206.07207
This work addresses scaling up the sketch classification task into a large number of categories. Collecting sketches for training is a slow and tedious process that has so far precluded any attempts to large-scale sketch recognition. We overcome the
Externí odkaz:
http://arxiv.org/abs/2202.13164
We introduce Object-Guided Localization (OGuL) based on a novel method of local-feature matching. Direct matching of local features is sensitive to significant changes in illumination. In contrast, object detection often survives severe changes in li
Externí odkaz:
http://arxiv.org/abs/2202.04445
Autor:
Papakipos, Zoë, Tolias, Giorgos, Jenicek, Tomas, Pizzi, Ed, Yokoo, Shuhei, Wang, Wenhao, Sun, Yifan, Zhang, Weipu, Yang, Yi, Addicam, Sanjay, Papadakis, Sergio Manuel, Ferrer, Cristian Canton, Chum, Ondrej, Douze, Matthijs
The 2021 Image Similarity Challenge introduced a dataset to serve as a new benchmark to evaluate recent image copy detection methods. There were 200 participants to the competition. This paper presents a quantitative and qualitative analysis of the t
Externí odkaz:
http://arxiv.org/abs/2202.04007