Pith. sign in

REVIEW 5 cited by

StackGAN++: Realistic Image Synthesis with Stacked Generative Adversarial Networks

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1710.10916 v3 pith:QBX4EMUG submitted 2017-10-19 cs.CV cs.AIstat.ML

Han Zhang , Tao Xu , Hongsheng Li , Shaoting Zhang , Xiaogang Wang , Xiaolei Huang , Dimitris Metaxas This is my paper · ORCID
classification cs.CVcs.AIstat.ML
keywords generativeadversarialimagesnetworksgeneratingmultiplephoto-realisticstacked
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Although Generative Adversarial Networks (GANs) have shown remarkable success in various tasks, they still face challenges in generating high quality images. In this paper, we propose Stacked Generative Adversarial Networks (StackGAN) aiming at generating high-resolution photo-realistic images. First, we propose a two-stage generative adversarial network architecture, StackGAN-v1, for text-to-image synthesis. The Stage-I GAN sketches the primitive shape and colors of the object based on given text description, yielding low-resolution images. The Stage-II GAN takes Stage-I results and text descriptions as inputs, and generates high-resolution images with photo-realistic details. Second, an advanced multi-stage generative adversarial network architecture, StackGAN-v2, is proposed for both conditional and unconditional generative tasks. Our StackGAN-v2 consists of multiple generators and discriminators in a tree-like structure; images at multiple scales corresponding to the same scene are generated from different branches of the tree. StackGAN-v2 shows more stable training behavior than StackGAN-v1 by jointly approximating multiple distributions. Extensive experiments demonstrate that the proposed stacked generative adversarial networks significantly outperform other state-of-the-art methods in generating photo-realistic images.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. AutoGAN: Neural Architecture Search for Generative Adversarial Networks

    cs.CV 2019-08 conditional novelty 7.0 of 10

    AutoGAN applies reinforcement-learning-based neural architecture search to GAN generators, discovering a CIFAR-10 architecture with FID 12.42 and an STL-10 FID 31.01, both state of the art in 2019.

  2. Dual Adversarial Inference for Text-to-Image Synthesis

    cs.CV 2019-08 conditional novelty 6.0 of 10

    A GAN for text-to-image synthesis that learns disentangled content and style codes via dual adversarial inference and cycle consistency, improving FID on Oxford-102, CUB, and COCO at 64x64.

  3. Sparse Generative Adversarial Network

    cs.CV 2019-08 conditional novelty 5.0 of 10

    A sparse-patch GAN with an encoder reconstructor reports improved Inception scores on CIFAR-10 and CelebA, but its mode-collapse guarantee is asserted, not proven.

  4. MemeFaceGenerator: Adversarial Synthesis of Chinese Meme-face from Natural Sentences

    cs.CL 2019-08 reject novelty 3.0 of 10

    A GAN-based system generates Chinese meme-face images from text by conditioning on an image template, but the supporting evidence is subjective and no baseline is provided.

  5. Systematic Analysis of Image Generation using GANs

    cs.LG 2019-08 conditional novelty 1.0 of 10

    The paper is a review that classifies GAN image-generation frameworks into text-to-image and image-to-image categories and compares them qualitatively.

Pith tools