A PPO-Based Bitrate Allocation Conditional Diffusion Model for Remote Sensing Image Compression

Yuming Han; Jooho Kim; Anish Shakya

cs.CVMar 16, 2026

A PPO-Based Bitrate Allocation Conditional Diffusion Model for Remote Sensing Image Compression

Authors: Yuming Han, Jooho Kim, Anish Shakya

Organizations: Department of Electrical and Computer Engineering, Texas A&M University, College Station, TX 77843 USA · Institute for a Disaster Resilient Texas, Texas A&M University, College Station, TX 77843 USA · Department of Marine & Coastal Environmental Science, Texas A&M University, Galveston, TX 77554 USA

Abstract

Existing remote sensing image compression methods still explore to balance high compression efficiency with the preservation of fine details and task-relevant information. Meanwhile, high-resolution drone imagery offers valuable structural details for urban monitoring and disaster assessment, but large-area datasets can easily reach hundreds of gigabytes, creating significant challenges for storage and long-term management. In this paper, we propose a PPO-based bitrate allocation Conditional Diffusion Compression (PCDC) framework. PCDC integrates a conditional diffusion decoder with a PPO-based block-wise bitrate allocation strategy to achieve high compression ratios while maintaining strong perceptual performance. We also release a high-resolution drone image dataset with richer structural details at a consistent low altitude over residential neighborhoods in coastal urban areas. Experimental results show compression ratios of 19.3x on DIV2K and 21.2x on the drone image dataset. Moreover, downstream object detection experiments demonstrate that the reconstructed images preserve task-relevant information with negligible performance loss.

Figures & tables

Fig. 1: System model of the proposed diffusion-based image compression framework with PPO-based bitrate allocation.

Fig. 2: Data collection points and sample images in Galveston, Texas.

Model	DIV2K	Drone Image Dataset
BPG	10.8	11.6
HiFiC	15.9	16.8
CDC	16.4	17.2
PCDC (Ours)	19.3	21.2

TABLE I: Compression ratio comparison on DIV2K and drone image datasets.

Fig. 3: Visual comparison of reconstruction images by different methods. The first two rows are from the proposed drone image dataset and the last row is from DIV2K. (a) BPG, (b) HiFiC, (c) CDC, and (d) the proposed PCDC. The proposed method reconstructs sharper edges and clearer structural details.

Fig. 4: Rate–distortion comparison of BPG, HiFiC, CDC, and the proposed PCDCin terms of LPIPS, DISTS, and PSNR across different bitrates (bpp).

Model	Data	B-R	V-R	Conf.	IoU
YOLO11s-pre	Ori.	–	0.99	V: 0.825	V: 0.97
YOLO11s-pre	Rec.	–	0.98	V: 0.828	V: 0.97
YOLO11s-ft	Ori.	1.02	0.96	B: 0.789, V: 0.743	B: 0.95, V: 0.96
YOLO11s-ft	Rec.	1.00	0.95	B: 0.789, V: 0.740	B: 0.95, V: 0.96
YOLO12s-pre	Ori.	–	0.96	V: 0.840	V: 0.97
YOLO12s-pre	Rec.	–	0.95	V: 0.840	V: 0.97

TABLE II: Detection results on original and reconstructed images.

Fig. 5: Visual comparisons with corresponding confidence scores for building (first row) and vehicle (second row) detections.

Learned Image Compression

Learned image compression uses neural encoders, quantization, and entropy models to reduce image bitrates while balancing fidelity, perceptual quality, and task utility.

See all 120 papers

Dual-Constrained Diffusion Image Compression for Operational Rate-Distortion-Perception Optimization
Jun 11, 2026Sanxin Jiang, Jiro Katto, Heming SunLearned Image Compression Rate-Distortion Optimization
Department of Information Engineering Shanghai University of Electric Power Shanghai, China · Department of Computer Science and Communication Engineering Waseda University Tokyo, Japan · Faculty of Engineering Institute of Science Tokyo Tokyo, Japan
Exploiting Semantic and Pixel Representations for Ultra-Low Bitrate Image Compression
Jun 1, 2026Hao Wei, Yanhui Zhou, Chenyang Ge +2Learned Image Compression Diffusion Models
National Key Laboratory of Human-Machine Hybrid Augmented Intelligence, Institute of Artificial Intelligence and Robotics, Xi’an Jiaotong University, Xi’an 710049, China · School of Information and Telecommunication, Xi’an Jiaotong University, Xi’an 710049, China · Department of Computer Science and Software Engineering, The University of Western Australia, Perth, Crawley, WA 6009, Australia
Scalable Differentially Private Data Compression via Diffusion and Stochastic Codes
Jul 3, 2026Gergely Flamich, Oykü Sıla Güner, Yanxiao Liu +1Learned Image Compression Differential Privacy
Imperial College London, London, UK

Jun 11, 2026J/K move · Enter open · S save

Dual-Constrained Diffusion Image Compression for Operational Rate-Distortion-Perception Optimization

Sanxin Jiang, Jiro Katto, Heming Sun

Department of Information Engineering Shanghai University of Electric Power Shanghai, China · Department of Computer Science and Communication Engineering Waseda University Tokyo, Japan · Faculty of Engineering Institute of Science Tokyo Tokyo, Japan

Learned Image Compression Rate-Distortion Optimization Distortion-Perception Tradeoff Generative Image Compression Diffusion Models

Open paper AskPDF arXiv

The rate-distortion-perception (RDP) trade-off extends classical rate--distortion theory by imposing a distributional constraint on reconstructions, providing a unified framework for neural image compression that jointly governs fidelity and perceptual realism. While prior work achieves near-optimal rate--perception trade-offs, practical frameworks explicitly realizing the full RDP surface remain scarce, primarily due to the difficulty of introducing common randomness at the decoder. We propose DCIC (Dual-Constrained Diffusion Image Compression), which integrates a learned codec with a diffusion-based decoder governed by joint distortion and idempotence constraints. The distortion constraint bounds reconstruction fidelity relative to the base codec output; the idempotence constraint -- requiring that re-encoding the restored image recovers the base codec reconstruction -- serves as a tractable surrogate for the distributional perception requirement. Together, they steer the reverse denoising process via iterative optimization with consistent noise injection, realizing common randomness without additional rate overhead. At fixed rate, dual attenuation factors

(K_D, K_P)

jointly navigate the Pareto frontier of the distortion-perception plane, enabling continuously adjustable fidelity-realism trade-offs from a single bitstream. DCIC

_{RD}

(

K_P{=}0

) and DCIC

_{RP}

(

K_D{=}0

) arise as boundary curves, with DCIC

_{RDP}

(

K_D = K_P=1

) realizing the optimal interior operating point. Experiments on CelebA-HQ, CLIC2020, and ImageNet-1K across CNN, Transformer, and hybrid architectures confirm that DCIC

_{RDP}

achieves superior BD-PSNR over all perceptual codecs, while DCIC

_{RP}

matches dedicated perception-oriented methods in BD-FID, validating the practical value of full RDP surface navigation.

Abstract

Learned Image Compression

Explore similar work

Dual-Constrained Diffusion Image Compression for Operational Rate-Distortion-Perception Optimization

Exploiting Semantic and Pixel Representations for Ultra-Low Bitrate Image Compression

Scalable Differentially Private Data Compression via Diffusion and Stochastic Codes