RECORD
T2CI-GAN: Text to Compressed Image generation using Generative Adversarial Network
Computer Vision and Pattern Recognition · Submitted 1 Oct 2022
abstract compiled · 2026-09-23 · no language-model review
The authors' main claim The proposed models are tested on an open source benchmark dataset Oxford-102 Flower images using both RGB and JPEG compressed versions, and accomplished the state-of-the-art performance in the JPEG compressed domain.
The compiled record
The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
Open the compiled record
The abstract, sentence by sentence
The problem of generating textual descriptions for the visual data has gained research attention in the recent years. In contrast to that the problem of generating visual data from textual descriptions is still very challenging, because it requires the combination of both Natural Language Processing (NLP) and Computer Vision techniques. The existing methods utilize the Generative Adversarial Networks (GANs) and generate the uncompressed images from textual description. However, in practice, most of the visual data are processed and transmitted in the compressed representation. Hence, the proposed work attempts to generate the visual data directly in the compressed representation form using Deep Convolutional GANs (DCGANs) to achieve the storage and computational efficiency. We propose GAN models for compressed image generation from text. The first model is directly trained with JPEG compressed DCT images (compressed domain) to generate the compressed images from text descriptions. The second model is trained with RGB images (pixel domain) to generate JPEG compressed DCT representation from text descriptions. The proposed models are tested on an open source benchmark dataset Oxford-102 Flower images using both RGB and JPEG compressed versions, and accomplished the state-of-the-art performance in the JPEG compressed domain. The code will be publicly released at GitHub after acceptance of paper.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | The problem of generating textual descriptions for the visual data has gained research attention in the recent years. | background 1.00 | 0.04 | 0.02 | asserted 1.00 |
| s1 | In contrast to that the problem of generating visual data from textual descriptions is still very challenging, because it requires the combination of both Natural Language Processing (NLP) and Computer Vision techniques. | gap 0.50 | 0.05 | 0.03 | asserted 1.00 |
| s2 | The existing methods utilize the Generative Adversarial Networks (GANs) and generate the uncompressed images from textual description. | background 0.98 | 0.03 | 0.03 | asserted 1.00 |
| s3 | However, in practice, most of the visual data are processed and transmitted in the compressed representation. | background 0.82 | 0.05 | 0.03 | asserted 0.98 |
| s4 | Hence, the proposed work attempts to generate the visual data directly in the compressed representation form using Deep Convolutional GANs (DCGANs) to achieve the storage and computational efficiency. | contribution 0.93 | 0.55 | 0.20 | asserted 0.73 |
| s5 | We propose GAN models for compressed image generation from text. | contribution 0.99 | 0.91 | 0.25 | asserted 1.00 |
| s6 | The first model is directly trained with JPEG compressed DCT images (compressed domain) to generate the compressed images from text descriptions. | method 0.96 | 0.91 | 0.07 | asserted 1.00 |
| s7 | The second model is trained with RGB images (pixel domain) to generate JPEG compressed DCT representation from text descriptions. | method 0.96 | 0.91 | 0.05 | asserted 1.00 |
| s8 | The proposed models are tested on an open source benchmark dataset Oxford-102 Flower images using both RGB and JPEG compressed versions, and accomplished the state-of-the-art performance in the JPEG compressed domain. | result 0.99 | 0.96 | 0.61 | asserted 1.00 |
| s9 | The code will be publicly released at GitHub after acceptance of paper. | future work 0.96 | 0.30 | 0.02 | asserted 0.95 |
What was examined, and what was not
- arXiv listing abstract v1
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
The paper
Original abstract
The problem of generating textual descriptions for the visual data has gained research attention in the recent years. In contrast to that the problem of generating visual data from textual descriptions is still very challenging, because it requires the combination of both Natural Language Processing (NLP) and Computer Vision techniques. The existing methods utilize the Generative Adversarial Networks (GANs) and generate the uncompressed images from textual description. However, in practice, most of the visual data are processed and transmitted in the compressed representation. Hence, the proposed work attempts to generate the visual data directly in the compressed representation form using Deep Convolutional GANs (DCGANs) to achieve the storage and computational efficiency. We propose GAN models for compressed image generation from text. The first model is directly trained with JPEG compressed DCT images (compressed domain) to generate the compressed images from text descriptions. The second model is trained with RGB images (pixel domain) to generate JPEG compressed DCT representation from text descriptions. The proposed models are tested on an open source benchmark dataset Oxford-102 Flower images using both RGB and JPEG compressed versions, and accomplished the state-of-the-art performance in the JPEG compressed domain. The code will be publicly released at GitHub after acceptance of paper.
Human review
How the literature uses it
What this paper does with the arXiv papers it cites
What it cites
Cites as background 2 papers
-
The Conditional GAN model [ref] makes use of some additional information along with the noise.
Conditional Generative Adversarial Nets · Generative Adversarial Network (GAN) · 0.98
-
Some of the prominent works in compressed document images are discussed in [ref] and [ref] .
Extraction of Projection Profile, Run-Histogram and Entropy Features Straight from Run-Length Compressed Text-Documents · Introduction · 0.96
Discussion
Know something about this paper? Leave a comment, or highlight any passage above to comment on it.