Pith. sign in

REVIEW 14 cited by

High-Fidelity Image Compression with Score-based Generative Models

Not yet reviewed by Pith; the record is open.

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 2305.18231 v3 pith:UA4V25FP submitted 2023-05-26 eess.IV cs.CVcs.LGstat.ML

classification eess.IVcs.CVcs.LGstat.ML
keywords modelscompressiondiffusiongenerativeimagescore-basedsuccesstext-to-image
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Despite the tremendous success of diffusion generative models in text-to-image generation, replicating this success in the domain of image compression has proven difficult. In this paper, we demonstrate that diffusion can significantly improve perceptual quality at a given bit-rate, outperforming state-of-the-art approaches PO-ELIC and HiFiC as measured by FID score. This is achieved using a simple but theoretically motivated two-stage approach combining an autoencoder targeting MSE followed by a further score-based decoder. However, as we will show, implementation details matter and the optimal design decisions can differ greatly from typical text-to-image models.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 14 Pith papers

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

  1. Progressive Compression with Universally Quantized Diffusion Models

    cs.LG 2024-12 conditional novelty 7.0 of 10

    A uniform-noise diffusion model plus universal quantization enables a single-model progressive image codec that is computationally tractable and competitive with classic and neural baselines.

  2. Next-Frame Decoding for Ultra-Low-Bitrate Image Compression with Video Diffusion Priors

    cs.CV 2026-03 conditional novelty 6.5 of 10

    Ultra-low-bitrate image decoding is cast as one-step next-frame prediction from a compact anchor using adapted video diffusion priors, yielding large perceptual bitrate savings versus DiffC.

  3. SSDD: Single-Step Diffusion Decoder for Efficient Image Tokenization

    cs.CV 2025-10 conditional novelty 6.0 of 10

    A GAN-free, single-step distilled diffusion decoder that reconstructs images from latent codes with better perceptual FID than KL-VAE at higher throughput.

  4. StableCodec: Taming One-Step Diffusion for Extreme Image Compression

    eess.IV 2025-06 conditional novelty 6.0 of 10

    A one-step diffusion codec that compresses noisy latents at 64x and decodes with a single denoising step, setting state-of-the-art FID, KID, and DISTS at ultra-low bitrates.

  5. Generative Image Compression by Estimating Gradients of the Rate-variable Feature Distribution

    eess.IV 2025-05 conditional novelty 6.0 of 10

    A single rate-variable generative compression model treats quantization as a forward corruption and reverses it with a two-step denoiser, outperforming prior generative codecs on perceptual quality benchmarks.

  6. Higher fidelity perceptual image and video compression with a latent conditioned residual denoising diffusion model

    eess.IV 2025-05 conditional novelty 6.0 of 10

    ResCDC combines a fidelity-optimized decoder with a latent-conditioned residual denoising diffusion model to improve PSNR by up to 2 dB over CDC while retaining similar LPIPS and FID.

  7. Ultra Lowrate Image Compression with Semantic Residual Coding and Compression-aware Diffusion

    cs.CV 2025-05 conditional novelty 6.0 of 10

    ResULIC combines residual text captions, optimized prompts, and bitrate-aware diffusion steps to reach state-of-the-art perceptual quality at ultra-low bitrates.

  8. Why Compress What You Can Generate? When GPT-4o Generation Ushers in Image Compression Fields

    cs.CV 2025-04 conditional novelty 6.0 of 10

    A structured raster-scan prompt fed to GPT-4o can reconstruct images at ultra-low bitrates (around 0.001 bpp) with competitive perceptual quality versus trained generative codecs, though only demonstrated on 10 images.

  9. UniMIC: Towards Universal Multi-modality Perceptual Image Compression

    eess.IV 2024-12 conditional novelty 6.0 of 10

    A single text-conditioned diffusion refiner improves perceptual quality (FID, LPIPS) of images from eight different base codecs and extends to unseen codecs.

  10. Is JPEG AI going to change image forensics?

    eess.IV 2024-12 conditional novelty 6.0 of 10

    JPEG AI compression adds upsampling-like artifacts that push pristine images into the deepfake class and blind splicing localization, an effect that is opposite to standard JPEG and stronger at low bitrates.

  11. Fast Training-free Perceptual Image Compression

    eess.IV 2025-06 conditional novelty 5.0 of 10

    A noise-then-denoise decoder with a pre-trained diffusion model turns any existing codec into a fast, training-free perceptual codec with a KL-divergence guarantee and 0.1-10s decoding.

  12. Generative Latent Coding for Ultra-Low Bitrate Image and Video Compression

    eess.IV 2025-05 conditional novelty 5.0 of 10

    Transform coding in a VQ-VAE latent space, instead of pixel space, yields 45% bitrate savings over MS-ILLM at equal FID for images and 65.3% DISTS-based bitrate savings over PLVC for video at ultra-low bitrates.

  13. Controllable Distortion-Perception Tradeoff Through Latent Diffusion for Neural Image Compression

    eess.IV 2024-12 conditional novelty 5.0 of 10

    A pretrained neural image codec can be augmented with a decoder-side latent diffusion module and a τ slider that interpolates between low-distortion and high-perception reconstructions without changing the bitstream.

  14. Synonymous Variational Inference for Perceptual Image Compression

    cs.IT 2025-05 reject novelty 4.0 of 10

    A synonym-set formulation of variational inference re-derives the rate-distortion-perception tradeoff and is demonstrated with a single progressive image codec.

Pith tools