Image-to-word transformation based on dividing and vector quantizing images with words

By Y. Mori, Hironobu Takahashi, R. Oka. 1999

Subfields: Computer Vision and Pattern Recognition, Multimodal Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Multimodal Learning

Citations in the corpus

Learning Transferable Visual Models From Natural Language Supervision (2021, methodology): “However, as mentioned in the introduction and approach section, methods of leveraging natural language descriptions in computer vision well predate the use of this specific term, especially for image retrieval (Mori et al., 1999) and object classification (Wang et al.”
Combined Scaling for Zero-shot Transfer Learning (2021, methodology): “Similar to CLIP and ALIGN, our work also learns the mapping between images and texts, which is related to many previous works, such as (Hironobu et al., 1999; Weston et al., 2010; Socher and Fei-Fei, 2010; Socher et al., 2013; Hodosh et al., 2013; Frome et al., 2013; Norouzi et al., 2013; Kiros et…”
LAION-5B: An open large-scale dataset for training next generation image-text models (2022, background): “Learning from multimodal data such as text, images, and audio is a longstanding research challenge in machine learning [31, 51, 56, 83, 86].”
RegionCLIP: Region-based Language-Image Pretraining (2021, background): “The most relevant work is learning visual representations from natural language, such as image tags [3, 8, 12, 25, 28] and text descriptions [11,23,43,53,62].”
Automatic multimedia cross-modal correlation discovery (2004, methodology): “For example, to associate words with images for automatic image captioning, people have proposed methods based on classifiers [15] or information retrieval techniques (relevance model [10] and user feedback [26]), as well as building association models (translation model [7]; hierarchical model [2, ”
Combined Scaling for Open-Vocabulary Image Classiﬁcation (2022, methodology): “Similar to CLIP and ALIGN, our work also learns the mapping between images and texts, which is related to many previous works, such as (Hironobu et al., 1999; Weston et al., 2010; Socher and Fei-Fei, 2010; Socher et al., 2013; Hodosh et al., 2013; Frome et al., 2013; Norouzi et al., 2013; Kiros et…”
Baselines for Image Annotation (2010, background): “Mori et al. (1999) was among the first to consider the co-occurrence view of the translation process where the annotation of a query image could be inferred from examples of regions-keyword associations.”
The SUN Attribute Database: Beyond Categories for Deeper Scene Understanding (2014, methodology): “Hironobu et al. (1999) counts the number of co-occurrences of image patch features and caption keywords to find correlations between keywords and features .”
A review on automatic image annotation techniques (2012, methodology): “[18], training images are divided into blocks and blocks are clustered.”
Semantic Annotation of Satellite Images Using Latent Dirichlet Allocation (2010, background): “[4] exploit a co-occurrence model in which they look at the cooccurrence of words with natural image tiles created using a regular grid.”
XRCE ’ s participation to ImagEval (2007, background): “They simply divide the image into subimages and assign all textual keywords to each sub-image.”
Automatic Caption Generation for News Images (2013, methodology): “This very naive model works well on an encyclopedia collection database (Mori et al., 1999).”
Every Picture Tells a Story: Generating Sentences from Images (2010, unknown): “The ﬁrst image annotation system is due to Mori et al. [2]; Duygulu et al. continued this tradition using models from machine translation [3].”
Automatic Image Annotation Using Auxiliary Text Information (2008, methodology): “The co-occurrence model (Mori et al., 1999) collects co-occurrence counts between words and image features and uses them to predict annotations for new images.”