Search Results for author: Weixin Li

Found 13 papers, 3 papers with code

Zero-Shot Scene Graph Generation via Triplet Calibration and Reduction

no code implementations7 Sep 2023 Jiankai Li, Yunhong Wang, Weixin Li

In our framework, a triplet calibration loss is first presented to regularize the representations of diverse triplets and to simultaneously excavate the unseen triplets in incompletely annotated training scene graphs.

Graph Generation Scene Graph Generation

CUR Transformer: A Convolutional Unbiased Regional Transformer for Image Denoising

1 code implementation journal 2023 Kang Xu, Weixin Li, Xia Wang, Xiaoyan Hu, Ke Yan, Xiaojie Wang, Xuan Dong

Based on the prior that, for each pixel, its similar pixels are usually spatially close, our insights are that (1) we partition the image into non-overlapped windows and perform regional self-attention to reduce the search range of each pixel, and (2) we encourage pixels across different windows to communicate with each other.

Image Denoising Jpeg Compression Artifact Reduction +1

MIEHDR CNN: Main Image Enhancement based Ghost-Free High Dynamic Range Imaging using Dual-Lens Systems

no code implementations AAAI Technical Track on Computer Vision I 2021 Xuan Dong, Xiaoyan Hu, Weixin Li, Xiaojie Wang;Yunhong Wang

In most of the related HDR imaging methods, the problem is usually solved by Multiple Images Merging, i. e. the final HDR image is fused from pixels of all the input LDR images.

Denoising Image Enhancement

Cycle-CNN for Colorization towards Real Monochrome-Color Camera Systems

1 code implementation AAAI Technical Track: Vision 2020 Xuan Dong, Weixin Li, Xiaojie Wang, Yunhong Wang

We present a new CNN model, named cycle CNN, which can directly use the real data from monochrome-color camera systems for training.


VLAD3: Encoding Dynamics of Deep Features for Action Recognition

no code implementations CVPR 2016 Yingwei Li, Weixin Li, Vijay Mahadevan, Nuno Vasconcelos

To account for long-range inhomogeneous dynamics, a VLAD descriptor is derived for the LDS and pooled over the whole video, to arrive at the final VLAD^3 representation.

Action Recognition Temporal Action Localization

Joint Image-Text News Topic Detection and Tracking with And-Or Graph Representation

no code implementations15 Dec 2015 Weixin Li, Jungseock Joo, Hang Qi, Song-Chun Zhu

The AOG embeds a context sensitive grammar that can describe the hierarchical composition of news topics by semantic elements about people involved, related places and what happened, and model contextual relationships between elements in the hierarchy.


Fidelity-Naturalness Evaluation of Single Image Super Resolution

no code implementations21 Nov 2015 Xuan Dong, Yu Zhu, Weixin Li, Lingxi Xie, Alex Wong, Alan Yuille

In this paper, we proposed to use both fidelity (the difference with original images) and naturalness (human visual perception of super resolved images) for evaluation.

Image Quality Assessment Image Super-Resolution

Ground-truth dataset and baseline evaluations for image base-detail separation algorithms

no code implementations21 Nov 2015 Xuan Dong, Boyan Bonev, Weixin Li, Weichao Qiu, Xianjie Chen, Alan Yuille

Base-detail separation is a fundamental computer vision problem consisting of modeling a smooth base layer with the coarse structures, and a detail layer containing the texture-like structures.

Multiple Instance Learning for Soft Bags via Top Instances

no code implementations CVPR 2015 Weixin Li, Nuno Vasconcelos

Under this formulation, both positive and negative bags are soft, in the sense that negative bags can also contain positive instances.

Multiple Instance Learning

Recognizing Activities via Bag of Words for Attribute Dynamics

no code implementations CVPR 2013 Weixin Li, Qian Yu, Harpreet Sawhney, Nuno Vasconcelos

A video sequence is decomposed into short-term segments, which are characterized by the dynamics of their attributes.

Activity Recognition

Recognizing Activities by Attribute Dynamics

no code implementations NeurIPS 2012 Weixin Li, Nuno Vasconcelos

The proposed method is shown to outperform similar classifiers derived from the kernel dynamic system (KDS) and state-of-the-art approaches for dynamics-based or attribute-based action recognition.

Action Recognition Temporal Action Localization

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