Search Results for author: Wen-Hsiao Peng

Found 20 papers, 8 papers with code

Transformer-based Variable-rate Image Compression with Region-of-interest Control

no code implementations18 May 2023 Chia-Hao Kao, Ying-Chieh Weng, Yi-Hsin Chen, Wei-Chen Chiu, Wen-Hsiao Peng

Our prompt generation networks generate content-adaptive tokens according to the input image, an ROI mask, and a rate parameter.

Image Compression

Hierarchical B-frame Video Coding Using Two-Layer CANF without Motion Coding

no code implementations CVPR 2023 David Alexandre, Hsueh-Ming Hang, Wen-Hsiao Peng

The rate-distortion performance of our scheme is slightly lower than that of the state-of-the-art learned B-frame coding scheme, B-CANF, but outperforms other learned B-frame coding schemes.

Video Compression

Content-Adaptive Motion Rate Adaption for Learned Video Compression

no code implementations13 Feb 2023 Chih-Hsuan Lin, Yi-Hsin Chen, Wen-Hsiao Peng

This paper introduces an online motion rate adaptation scheme for learned video compression, with the aim of achieving content-adaptive coding on individual test sequences to mitigate the domain gap between training and test data.

Video Compression

HuPR: A Benchmark for Human Pose Estimation Using Millimeter Wave Radar

1 code implementation22 Oct 2022 Shih-Po Lee, Niraj Prakash Kini, Wen-Hsiao Peng, Ching-Wen Ma, Jenq-Neng Hwang

In addition to the benchmark, we propose a cross-modality training framework that leverages the ground-truth 2D keypoints representing human body joints for training, which are systematically generated from the pre-trained 2D pose estimation network based on a monocular camera input image, avoiding laborious manual label annotation efforts.

2D Pose Estimation Pose Estimation

B-CANF: Adaptive B-frame Coding with Conditional Augmented Normalizing Flows

no code implementations5 Sep 2022 Mu-Jung Chen, Yi-Hsin Chen, Peng-Yu Chen, Chih Hsuan Lin, Yung-Han Ho, Wen-Hsiao Peng

This work introduces a B-frame coding framework, termed B-CANF, that exploits conditional augmented normalizing flows for B-frame coding.

CANF-VC: Conditional Augmented Normalizing Flows for Video Compression

1 code implementation12 Jul 2022 Yung-Han Ho, Chih-Peng Chang, Peng-Yu Chen, Alessandro Gnutti, Wen-Hsiao Peng

CANF-VC represents a new attempt that leverages the conditional ANF to learn a video generative model for conditional inter-frame coding.

Video Compression

ANFIC: Image Compression Using Augmented Normalizing Flows

1 code implementation18 Jul 2021 Yung-Han Ho, Chih-Chun Chan, Wen-Hsiao Peng, Hsueh-Ming Hang, Marek Domanski

This paper introduces an end-to-end learned image compression system, termed ANFIC, based on Augmented Normalizing Flows (ANF).

Image Compression

A Dual-Critic Reinforcement Learning Framework for Frame-level Bit Allocation in HEVC/H.265

no code implementations5 Apr 2021 Yung-Han Ho, Guo-Lun Jin, Yun Liang, Wen-Hsiao Peng, Xiaobo Li

This paper introduces a dual-critic reinforcement learning (RL) framework to address the problem of frame-level bit allocation in HEVC/H. 265.

reinforcement-learning Reinforcement Learning (RL)

Weakly-Supervised Image Semantic Segmentation Using Graph Convolutional Networks

1 code implementation31 Mar 2021 Shun-Yi Pan, Cheng-You Lu, Shih-Po Lee, Wen-Hsiao Peng

One common approach to this task is to propagate the activation scores of Class Activation Maps (CAMs) using a random-walk mechanism in order to arrive at complete pseudo labels for training a semantic segmentation network in a fully-supervised manner.

Weakly-Supervised Semantic Segmentation

Video Rescaling Networks with Joint Optimization Strategies for Downscaling and Upscaling

1 code implementation CVPR 2021 Yan-Cheng Huang, Yi-Hsin Chen, Cheng-You Lu, Hui-Po Wang, Wen-Hsiao Peng, Ching-Chun Huang

Our Long Short-Term Memory Video Rescaling Network (LSTM-VRN) leverages temporal information in the low-resolution video to form an explicit prediction of the missing high-frequency information for upscaling.

GSVNet: Guided Spatially-Varying Convolution for Fast Semantic Segmentation on Video

1 code implementation16 Mar 2021 Shih-Po Lee, Si-Cun Chen, Wen-Hsiao Peng

Moreover, we introduce a guided spatially-varying convolution for fusing segmentations derived from the previous and current frames, to mitigate propagation error and enable lightweight feature extraction on non-keyframes.

Image Segmentation Semantic Segmentation +2

Learning Priors for Adversarial Autoencoders

no code implementations ICLR 2018 Hui-Po Wang, Wen-Hsiao Peng, Wei-Jan Ko

Most deep latent factor models choose simple priors for simplicity, tractability or not knowing what prior to use.

Image Generation Translation

All about Structure: Adapting Structural Information across Domains for Boosting Semantic Segmentation

1 code implementation CVPR 2019 Wei-Lun Chang, Hui-Po Wang, Wen-Hsiao Peng, Wei-Chen Chiu

In this paper we tackle the problem of unsupervised domain adaptation for the task of semantic segmentation, where we attempt to transfer the knowledge learned upon synthetic datasets with ground-truth labels to real-world images without any annotation.

Semantic Segmentation Synthetic-to-Real Translation +2

An Autoencoder-based Learned Image Compressor: Description of Challenge Proposal by NCTU

no code implementations20 Feb 2019 David Alexandre, Chih-Peng Chang, Wen-Hsiao Peng, Hsueh-Ming Hang

We propose a lossy image compression system using the deep-learning autoencoder structure to participate in the Challenge on Learned Image Compression (CLIC) 2018.

Image Compression MS-SSIM +1

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