REVIEW 8 cited by
Lossy Compression with Gaussian Diffusion
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
read the original abstract
We consider a novel lossy compression approach based on unconditional diffusion generative models, which we call DiffC. Unlike modern compression schemes which rely on transform coding and quantization to restrict the transmitted information, DiffC relies on the efficient communication of pixels corrupted by Gaussian noise. We implement a proof of concept and find that it works surprisingly well despite the lack of an encoder transform, outperforming the state-of-the-art generative compression method HiFiC on ImageNet 64x64. DiffC only uses a single model to encode and denoise corrupted pixels at arbitrary bitrates. The approach further provides support for progressive coding, that is, decoding from partial bit streams. We perform a rate-distortion analysis to gain a deeper understanding of its performance, providing analytical results for multivariate Gaussian data as well as theoretic bounds for general distributions. Furthermore, we prove that a flow-based reconstruction achieves a 3 dB gain over ancestral sampling at high bitrates.
Forward citations
Cited by 8 Pith papers
-
Learning-to-Defer with Expert-Conditional Advice
The paper defines Learning-to-Defer with advice, proves inconsistency of separated surrogate losses, and introduces an augmented surrogate with H-consistency and excess-risk bounds that recovers the Bayes-optimal policy.
-
Remote Channel Synthesis
The paper gives a single-letter characterization of optimal rates for remote channel synthesis and proves that direct per-symbol synthesis is strictly suboptimal at low common randomness.
-
Next-Frame Decoding for Ultra-Low-Bitrate Image Compression with Video Diffusion Priors
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.
-
GVCCTurbo: Rate-Compute Quality Scheduling for Codebook Driven Generative Compression
GVCCTurbo reuses cached generator endpoints across codebook correction steps, cutting denoiser evaluations from 20 to 9 and decoding time by roughly 44% at similar LPIPS.
-
Scalable Differentially Private Data Compression via Diffusion and Stochastic Codes
Step-limited PPR plus Laplace DiffC yields a pure-LDP image compressor that cuts bitrate 10–30× versus privatize-then-compress on CIFAR-10 classification.
-
StableCodec: Taming One-Step Diffusion for Extreme Image Compression
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.
-
Fast Training-free Perceptual Image Compression
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.
-
Synonymous Variational Inference for Perceptual Image Compression
A synonym-set formulation of variational inference re-derives the rate-distortion-perception tradeoff and is demonstrated with a single progressive image codec.
Discussion (0). Continue with ORCID to comment.