News

MCL Research on Texture Synthesis

Automatic   synthesis   of   visually   pleasant   texture   that resembles  exemplary  texture  finds  applications  in  computer  graphics.  Texture  synthesis  has  been  studied  for several  decades  since  it  is  also  of  theoretical  interest  in texture analysis and modeling. Texture can be synthesized pixel-by-pixel or patch-by-patch based on an exemplary  pattern.  For  the  pixel-based  synthesis,  a  pixel conditioned on its squared neighbor was synthesized using the  conditional  probability  and  estimated  by  a  statistical method. Generally,  patch-based  texture  synthesis yields  higher  quality  than  pixel-based  texture  synthesis. Yet, searching the whole image for patch-based synthesis is  extremely  slow.  To  speed  up  the  process,  small patches of the exemplary texture can be stitched together to  form  a  larger  region. Although  these  methods can produce texture of higher quality, the diversity of produced textures is limited. Besides texture synthesis in the spatial domain, texture images from the spatial domain can be transformed to the spectral domain with certain filters (or kernels), thus exploiting the statistical correlation of filter responses for texture synthesis. Commonly used kernels include the Gabor filters and the steerable pyramid filter banks.

We  have  witnessed  amazing  quality  improvement  of synthesized  texture  over  the  last  five  to  six  years  due to  the  resurgence  of  neural  networks.  Texture  synthesis based on deep learning (DL), such as Convolutional Neural Networks  (CNNs)  and  Generative  Adversarial  Networks(GANs), yield visually pleasing results. DL-based methods learn transform kernels from numerous training data through end-to-end optimization. However, these methods have two main shortcomings: 1) a lack of mathematical  transparency  and  2)  a  higher  training  and  inference complexity. To address these drawbacks, we investigate a non-parametric and interpretable texture synthesis method, called NITES [1].

NITES  consists  of  three  steps.  First,  it  analyzes  the exemplary texture to obtain its joint spatial-spectral [...]

By |June 27th, 2021|News|Comments Off on MCL Research on Texture Synthesis|

MCL Research on Image Denoising

As a fundamental Computer vision problem, image denoising aims at reducing noise images to improve resolutions.  As a sub-topic of image restoration, image denoising not only has wide applications in practical problems, but also can be important pre-processing procedures for other CV or NLP problems.

Traditionally, algorithms by patch-wise denoising, like Non-local Mean and BM3D, usually assume the noise are Gaussian noise try to reduce noise by the randomness of noise and signal preservation across similar patches. After CNN architecture introduced to CV field, similar to other image restoration problems like super-resolution, deblurring, and dehazing, denoising problem also developed out CNN-based methods, with two main streams: one focus more on pixel-wise restoration and the other cares more about overall pleasure. Besides, combining different image restoration problems together, that building a more general model which can work on multiple image restoration problems gradually obtained more attention.

With better performance achieved in denoising problem, more and more algorithms suffer from the model size and reference speed. We would like to introduce SSL principle to tackle denoising problem with comparable performance while with higher efficiency in the future.

Image credit: Dabov, Kostadin, et al. “Image denoising by sparse 3-D transform-domain collaborative filtering.” IEEE Transactions on image processing 16.8 (2007): 2080-2095.

By |June 21st, 2021|News|Comments Off on MCL Research on Image Denoising|

Welcome MCL New Member – Xinyu Wang

In Summer 2021, we have a new MCL member, Xinyu Wang, joining our big family. Here is a short interview with Xinyu with our great welcome.

1. Could you briefly introduce yourself and your research interests?

My name is Xinyu Wang, I am a Master student in Electrical Engineering, and it’s my second year at USC. I am new here as a summer intern. My research interests mainly include machine learning and robotics, and I will work on image forensics related topics this summer at MCL.

2. Could you briefly introduce yourself and your research interests?

MCL has a group of motivated and intelligent people, who are full of passion about their research. And I am impressed by the open and friendly atmosphere here. People are encouraged to show their ideas and help each other, and everyone is friendly and supportive.

3. What is your future expectation and plan in MCL?

This summer, I am working with Yao on image forensics topics using the green learning method, under the supervision of Professor Kuo. I believe this will be a great opportunity for me to further explore machine learning and working on this interesting topic. I also hope to make new friends here and make lasting connections with MCL members.

By |June 13th, 2021|News|Comments Off on Welcome MCL New Member – Xinyu Wang|

Welcome MCL New Member – Peida Han

In Summer 2021, we have a new MCL member, Peida Han, joining our big family. Here is a short interview with Peida with our great welcome.

1. Could you briefly introduce yourself and your research interests?

My name is Peida Han, and I am a first year master student in Computer Science (artificial intelligence) at USC. I received my Bachelor’s degree in Computer Science and Engineering from the Ohio State University in 2016. I previously worked on some machine learning based projects such as an autonomous aerial system on drones in my undergrad. With strong interest in image processing for practical real life applications, I am currently working on the Breast Cancer Image segmentation project in MCL.

2. What is your impression about MCL and USC?

My impression about MCL is that the lab members are friendly and motivating. I feel everyone is approachable and the whole group like to help each other out. In addition, everyone is dedicated to their work and I am inspired to work hard and learn from them. USC provides valuable resources from the perspectives of both academia and industry. My impression of USC is that students have access to resources easily and professors have high standards of course quality, and there are many other valuable resources. I am glad to be a part of the Trojan family.

3. What is your future expectation and plan in MCL?

My expectation in MCL is to explore my potentials in pure research, especially in the image processing field. And I am glad I can be involved in the research of image segmentation and hope that could be helpful to society. I learnt great a lot from Prof. Kuo and I hope I can contribute my own efforts in MCL. It [...]

By |June 6th, 2021|News|Comments Off on Welcome MCL New Member – Peida Han|

MCL Research on SSL-based Image Anomaly Localization

Image anomaly localization is an important problem in image processing and computer vision, with numerous applications in many areas, such as industrial manufacturing inspection, medical image diagnosis and even video surveillance analysis. The goal of image anomaly localization is to locate the anomaly or anomalous region on the pixel level. Like most other anomaly detection problems, we formulate image anomaly localization as an unsupervised task. More specifically, it means training set only contains normal images, and no anomalous images and corresponding labeled masks are available during model training. This is because anomalous examples are either too expensive to collect or too few to be modeled, which makes it an extremely challenging yet attracting problem.

To tackle this problem, we propose a new image anomaly localization method, called AnomalyHop [1], based on the successive subspace learning (SSL) framework. This is also the first work that applies SSL to the anomaly localization problem. AnomalyHop consists of three modules: 1) feature extraction via successive subspace learning (SSL), 2) normality feature distributions modeling via various Gaussian models, and 3) anomaly map generation and fusion. As compared with previous deep-learning-based image anomaly localization methods, AnomalyHop is mathematically transparent, easy to train, and fast in its inference speed. Besides that, its area under the ROC curve (ROC-AUC) performance on the MVTec AD dataset is 95.9%, which is the state-of-the-art performance.

-By Kaitai Zhang and Bin Wang

 

[1] Zhang, K., Wang, B., Wang, W., Sohrab, F., Gabbouj, M., & Kuo, C. C. J. (2021). AnomalyHop: An SSL-based Image Anomaly Localization Method. arXiv preprint arXiv:2105.03797.

By |May 31st, 2021|News|Comments Off on MCL Research on SSL-based Image Anomaly Localization|

Congratulations to Kaitai Zhang for Passing His Defense

Congratulations to Kaitai Zhang for passing his defense on May 19, 2021. His Ph.D. thesis is entitled “Data-Driven Image Analysis, Modeling, Synthesis and Anomaly Localization Techniques”. Here we invite Kaitai to share a brief introduction of his thesis and some words he would like to say at the end of the Ph.D. study journey.

1) Abstract of Thesis

Emerging Deep learning and machine learning techniques have brought impressive improvements for numerous topics in image processing and computer vision fields. In this thesis, we introduce our research on Data-Driven Image Analysis, Modeling, Synthesis and Anomaly Localization Techniques: 1) image anomaly detection and localization; 2) texture analysis, modeling and synthesis.

For the first part, we will focus on image anomaly detection and localization tasks. Image anomaly detection is a binary classification problem to determine whether an input contains an anomaly, and image anomaly localization is to get pixel-precise segmentation of regions that appear anomalous. Detecting and localizing anomalies is a critical and long-standing problem in image processing and computer vision, and has applications in many real-world scenarios like medical image diagnosis and automated manufacturing inspection. In this talk, I will introduce two of our recent works, PEDENet and AnomalyHop. PEDENet is a neural network-based framework that jointly learns image local feature and density estimation model. AnomalyHop employs successive subspace learning (SSL) framework, and utilizes various Gaussian Descriptors to learn normality feature distributions. Both of them achieve state-of-the-art performance on MVTec AD dataset, and provide either smaller model size or faster inference speed.

In the second part, our previous works in texture analysis, modeling and synthesis will be reviewed. For dynamic texture synthesis, two effective techniques will be proposed and proved effective. The enhanced model could encode coherence of local features as well as the [...]

By |May 24th, 2021|News|Comments Off on Congratulations to Kaitai Zhang for Passing His Defense|

MCL Research on Image Super-resolution

Image super-resolution (SR) is a classic image reconstruction problem in computer vision (CV), which aims at recovering a high-resolution image from a low-resolution image. As a type of supervised generative problem, image SR attracts wide attention due to its strong connection with other CV topics, such as object recognition, object alignment, texture synthesis and so on. Besides, it has extensive applications in real world, for example, medical diagnosis, remote sensing, biometric information identification, etc.

For the state-of-the-art approaches for SR, typically there are two mainstreams: 1) example-based learning methods, and 2) Deep Learning (CNN-based) methods. Example-based methods either exploit external low-high resolution exemplar pairs [1], or learn internal similarity of the same image with different resolution scales [2]. However, features used in example-based methods are usually traditional gradient-related or just handcraft, which may affect model performance. While CNN-based SR methods (e.g. SRCNN [3]) does not really distinguish between feature extraction and decision making. Lots of basic CNN models/blocks are applied to SR problem, e.g. GAN, residual learning, attention network, and provide superior SR results. Nevertheless, the non-explainable process and exhaustive training cost are serious drawbacks of CNN-based methods.

We propose a Successive-Subspace-Learning-based (SSL-based) method to gradually partition data into subspaces by feature statistics. In addition, we utilize spatial-spectral compatible cw-Saab features to express exemplar pairs by taking advantage of reasonable feature extraction [4]. In the future, we aim at providing such a SSL-based explainable method with high efficiency for SR problem.

—  By Wei Wang

 

Reference:

[1] Timofte, Radu, Vincent De Smet, and Luc Van Gool. “A+: Adjusted anchored neighborhood regression for fast super-resolution.” Asian conference on computer vision. Springer, Cham, 2014.

[2] Huang, Jia-Bin, Abhishek Singh, and Narendra Ahuja. “Single image super-resolution from transformed self-exemplars.” Proceedings of the IEEE conference on [...]

By |May 16th, 2021|News|Comments Off on MCL Research on Image Super-resolution|

MCL Research on Speed-up of Multi-Class XGBoost Classifier

Machine learning has witnessed a rapid increase in the amount of training/testing data, feature dimensions and class number due to the arrival of the big data era. In many applications, the systems are expected to deal with a very large number of classes and a huge amount of training/test data. These impose major challenges in: 1) classification accuracy, 2) model complexity in terms of the number of model parameters, and 3)  computational complexity in terms of training and testing costs.  Although deep-learning-based (DL-based) systems can provide good performance in many application contexts, their model sizes are large and training complexities are high.

A popular machine learning tool is the XGBoost [1], which is able to achieve excellent performance on many tasks. XGBoost is a boosting algorithm, which combines multiple weak classifiers to form a more powerful classifier. The XGBoost adds a tree in each iteration, which models the residue from the last iteration. We can simply sum the output from each tree to obtain the final prediction. For classification, XGBoost supports both binary classification and the multiclass prediction. However, multiclass classification tasks with XGBoost can take a very long time if the class number is huge. Hence, in our research, we aim to speed up the multiclass XGBoost.

 

Image credits:

[1] Chen, T., Guestrin, C., 2016.   Xgboost:  A scalable tree boosting system, in: Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining, pp. 785–794.

 

 

By |May 9th, 2021|News|Comments Off on MCL Research on Speed-up of Multi-Class XGBoost Classifier|
  • Permalink Gallery

    Professor Kuo Received 2021 IEEE CASS Charles A. Desoer Technical Achievement Award

Professor Kuo Received 2021 IEEE CASS Charles A. Desoer Technical Achievement Award

Congratulations to MCL Director, Professor C.-C. Jay Kuo, for being selected as the recipient of the 2021 IEEE CASS Charles A. Desoer Technical Achievement Award. This award is named after  Charles A. Desoer, Professor of Electrical Engineering and Computer Science, Emeritus, at UC Berkeley. This award honors individuals whose exceptional technical contributions to a field within the scope of the Circuits and Systems Society have been consistently evident over a period of years. Contributions are documented by publications and based on originality and continuity of effort. Professor Kuo received this award for his contributions to visual communications and multimedia systems

The 2021 CAS Society Awards Ceremony will be held virtually in parallel with the ISCAS 2021 event on Monday, 24 May. The ceremony will be pre-recorded and presented on both the online platform and at the live banquet in Daegu. Here is Professor Kuo’s acceptance speech.

“It is my great honor to receive the 2021 IEEE CASS Charles A. Desoer Technical Achievement Award. I know that this is a highly competitive award, and there are many well qualified nominees each year. I would like to give my deepest appreciation to the Technical Achievement Award Sub-Committee and the CASS Board for their recognition.

I am also grateful for the excellent research environment provided by the University of Southern California. It is a privilege to supervise 160 hardworking PhD students at USC in the last 30 years. We brainstorm research ideas, share research frustration, and enjoy research breakthrough together. I can say that there is nothing more rewarding than working with a large number of talented young people.

I have been heavily involved in two CASS technical committees in my career. They are the visual signal processing and communication technical [...]

By |May 2nd, 2021|News|Comments Off on Professor Kuo Received 2021 IEEE CASS Charles A. Desoer Technical Achievement Award|

MCL Research on Next Generation Video Coding

With the development of camera and sensor technologies, high-resolution images and videos have become ubiquitous in daily life. Demands on fast transmission and efficiency store high-quality images and videos increase dramatically. Problem on how to transmit and store media data efficiently have been widely discussed. Online high-resolution video meeting/live broadcasting also raise the pressure on fast encoding and decoding under the limitation of current bandwidth.

Numerous codecs have been developed during the past 20 years including the well know H.264, MPEG-4 and the latest H.265/HEVC which are widely used in our daily life. H26x and MPEG-x standards are well supported in both software and hardware. There are many encoder and decoder chipsets available commercially (for example chips from System on Chip Technologies Inc.) which can speed up the process and be configured based on user specifications. While the royal free codecs like AV1, it has higher complexity and not widely supported by the hardware chips which hinder it’s being widely used.

Previous frameworks used one layer transformation to perform the energy compaction task. We propose to use the multi-hop transform to perform this task which is supposed to have better energy compaction result. Presently, we focus on image compression (intra coding in video compression) to increase the performance while lower the complexity. In this project, we are trying to develop some low-complexity compression tools which can achieve comparable performance against the current standard.

By |April 25th, 2021|News|Comments Off on MCL Research on Next Generation Video Coding|