Search Results for author: Heming Sun

Found 25 papers, 8 papers with code

Tell Codec What Worth Compressing: Semantically Disentangled Image Coding for Machine with LMMs

no code implementations16 Aug 2024 Jinming Liu, Yuntao Wei, Junyan Lin, Shengyang Zhao, Heming Sun, Zhibo Chen, Wenjun Zeng, Xin Jin

We present a new image compression paradigm to achieve ``intelligently coding for machine'' by cleverly leveraging the common sense of Large Multimodal Models (LMMs).

Common Sense Reasoning Image Classification +5

Attack and Defense Analysis of Learned Image Compression

no code implementations18 Jan 2024 Tianyu Zhu, Heming Sun, Xiankui Xiong, Xuanpeng Zhu, Yong Gong, Minge jing, Yibo Fan

Our experiments compare the effects of different dimensions such as attack methods, models, qualities, and targets, concluding that in the worst case, there is a 61. 55% decrease in PSNR or a 19. 15 times increase in bpp under the PGD attack.

Image Compression

Accelerating Learnt Video Codecs with Gradient Decay and Layer-wise Distillation

no code implementations5 Dec 2023 Tianhao Peng, Ge Gao, Heming Sun, Fan Zhang, David Bull

In recent years, end-to-end learnt video codecs have demonstrated their potential to compete with conventional coding algorithms in term of compression efficiency.

Decoder Video Compression

SCP: Spherical-Coordinate-based Learned Point Cloud Compression

no code implementations24 Aug 2023 Ao Luo, Linxin Song, Keisuke Nonaka, Kyohei Unno, Heming Sun, Masayuki Goto, Jiro Katto

In recent years, the task of learned point cloud compression has gained prominence.

Prompt-ICM: A Unified Framework towards Image Coding for Machines with Task-driven Prompts

no code implementations4 May 2023 Ruoyu Feng, Jinming Liu, Xin Jin, Xiaohan Pan, Heming Sun, Zhibo Chen

For ICM, developing a unified codec to reduce information redundancy while empowering the compressed features to support various vision tasks is very important, which inevitably faces two core challenges: 1) How should the compression strategy be adjusted based on the downstream tasks?

Learned Image Compression with Mixed Transformer-CNN Architectures

2 code implementations CVPR 2023 Jinming Liu, Heming Sun, Jiro Katto

Most existing LIC methods are Convolutional Neural Networks-based (CNN-based) or Transformer-based, which have different advantages.

Image Compression

Multistage Spatial Context Models for Learned Image Compression

1 code implementation18 Feb 2023 Fangzheng Lin, Heming Sun, Jinming Liu, Jiro Katto

The proposed method features a comparable decoding speed to Checkerboard while reaching the RD performance of Autoregressive and even also outperforming Autoregressive.

Image Compression

ABCAS: Adaptive Bound Control of spectral norm as Automatic Stabilizer

no code implementations12 Nov 2022 Shota Hirose, Shiori Maki, Naoki Wada, Heming Sun, Jiro Katto

Spectral Normalization is one of the best methods for stabilizing the training of Generative Adversarial Network.

Generative Adversarial Network

Semantic Segmentation in Learned Compressed Domain

no code implementations3 Sep 2022 Jinming Liu, Heming Sun, Jiro Katto

Most machine vision tasks (e. g., semantic segmentation) are based on images encoded and decoded by image compression algorithms (e. g., JPEG).

Image Compression Segmentation +1

Streaming-capable High-performance Architecture of Learned Image Compression Codecs

1 code implementation2 Aug 2022 Fangzheng Lin, Heming Sun, Jiro Katto

Learned image compression allows achieving state-of-the-art accuracy and compression ratios, but their relatively slow runtime performance limits their usage.

Decoder Image Compression +1

Memory-Efficient Learned Image Compression with Pruned Hyperprior Module

no code implementations21 Jun 2022 Ao Luo, Heming Sun, Jinming Liu, Jiro Katto

Learned Image Compression (LIC) gradually became more and more famous in these years.

Image Compression

Q-LIC: Quantizing Learned Image Compression with Channel Splitting

no code implementations28 May 2022 Heming Sun, Lu Yu, Jiro Katto

Learned image compression (LIC) has reached a comparable coding gain with traditional hand-crafted methods such as VVC intra.

Image Compression MS-SSIM +2

End-to-End Learned Image Compression with Quantized Weights and Activations

no code implementations17 Nov 2021 Heming Sun, Lu Yu, Jiro Katto

To our best knowledge, this is the first work to give a complete analysis on the coding gain and the memory cost for a quantized LIC network, which validates the feasibility of the hardware implementation.

Image Compression MS-SSIM +2

Learned Image Compression with Separate Hyperprior Decoders

no code implementations31 Oct 2021 Zhao Zan, Chao Liu, Heming Sun, Xiaoyang Zeng, Yibo Fan

Learned image compression techniques have achieved considerable development in recent years.

Decoder Image Compression +2

Learned Video Compression with Residual Prediction and Loop Filter

1 code implementation19 Aug 2021 Chao Liu, Heming Sun, Jiro Katto, Xiaoyang Zeng, Yibo Fan

To reduce the complexity, a light ResNet structure is used as the backbone for both RP-Net and LF-Net.

Motion Compensation Video Compression

Fully Neural Network Mode Based Intra Prediction of Variable Block Size

1 code implementation5 Aug 2021 Heming Sun, Lu Yu, Jiro Katto

As far as we know, this is the first work to explore a fully NM based framework for intra prediction, and we reach a better coding gain with a lower complexity compared with the previous work.

regression

A QP-adaptive Mechanism for CNN-based Filter in Video Coding

no code implementations25 Oct 2020 Chao Liu, Heming Sun, Jiro Katto, Xiaoyang Zeng, Yibo Fan

Convolutional neural network (CNN)-based filters have achieved great success in video coding.

Quantization

COUGH: A Challenge Dataset and Models for COVID-19 FAQ Retrieval

1 code implementation EMNLP 2021 Xinliang Frederick Zhang, Heming Sun, Xiang Yue, Simon Lin, Huan Sun

For evaluation, we introduce Query Bank and Relevance Set, where the former contains 1, 236 human-paraphrased queries while the latter contains ~32 human-annotated FAQ items for each query.

16k Retrieval

Dual Learning-based Video Coding with Inception Dense Blocks

no code implementations22 Nov 2019 Chao Liu, Heming Sun, Junan Chen, Zhengxue Cheng, Masaru Takeuchi, Jiro Katto, Xiaoyang Zeng, Yibo Fan

This method is mainly composed of two parts: intra prediction and reconstruction filtering.

Learning Image and Video Compression through Spatial-Temporal Energy Compaction

no code implementations CVPR 2019 Zhengxue Cheng, Heming Sun, Masaru Takeuchi, Jiro Katto

Experimental results demonstrate that our proposed image compression outperforms the latest image compression standard with MS-SSIM quality metric, and provides higher performance compared with state-of-the-art learning compression methods at high bit rates, which benefits from our spatial energy compaction approach.

Decoder Image Compression +3

Cannot find the paper you are looking for? You can Submit a new open access paper.