Correcting Diffusion-Based Perceptual Image Compression with Privileged End-to-End Decoder
The proposed method leverages a privileged end-to-end decoder to correct the score function of a diffusion model, achieving better perceptual quality while guaranteeing distortion.
Theoretical Bound-Guided Hierarchical VAE for Neural Image Codecs Analysis
Utilizing theoretical bounds to guide neural image codecs improves performance significantly.
Neural Image Compression with Quantization Rectifier: Enhancing Image Quality Through Predictive Correction
Neural image compression is improved by the Quantization Rectifier method, preserving feature expressiveness for better image quality.
TLIC: Learned Image Compression with ROI-Weighted Distortion and Bit Allocation
Learned Image Compression with ROI-Weighted Distortion and Bit Allocation enhances perceptual quality through region-based bit allocation.
Laplacian-guided Entropy Model in Neural Codec with Blur-dissipated Synthesis: Enhancing Image Compression Quality
Non-isotropic diffusion model and innovative entropy model improve image compression quality.
Laplacian-guided Entropy Model in Neural Codec with Blur-dissipated Synthesis
Enhancing image compression through a conditional diffusion-based decoder and an innovative entropy model.
FTIC: Frequency-Aware Transformer for Learned Image Compression at ICLR 2024
提案された周波数感知トランスフォーマー（FAT）ブロックは、LICにおける多様な周波数情報のモデリングに挑戦し、状態-of-the-artのレート-歪み性能を達成します。
Super-High-Fidelity Image Compression with Hierarchical-ROI and Adaptive Quantization
Combining MSE-based models and generative models using Hierarchical-ROI and adaptive quantization improves visual quality at low bit rates.
Frequency-Aware Transformer for Learned Image Compression at ICLR 2024
Proposing a novel Frequency-Aware Transformer (FAT) block for learned image compression to capture multiscale and directional frequency components efficiently.
Achieving Realistic Image Compression at Ultra-Low Bitrates
Decoding images with iterative diffusion models can achieve realistic reconstructions at ultra-low bitrates, surpassing traditional codecs.
© 2024 by Linnk AI