• Permalink Gallery

    MCL member Vasileios Magoulianitis presented in the ICASSP 2023 conference

MCL member Vasileios Magoulianitis presented in the ICASSP 2023 conference

Vasileios had a trip to Greece to attend ICASSP 2023 and present two posters of MCL. Let’s hear what he would like to share about his experience:

ICASSP this year was held on the island of Rhodes in Greece which is popular touristic summer destination. The venue hosted a quite large number of presentations this year -most of them as posters-, interesting keynote speeches and other IEEE community side events, such as the celebration of Signal Processing Society (SPS) 75th anniversary.

I presented two works of our lab in form of posters, the one on rotation invariant 3D Point cloud classification, proposed by Pranav Kadam et al. and the new classification method from our lab, named  SLM (Classification via Subspace Learning Machine), proposed by Hongyu Fu. Most of the individuals that stopped by the posters, commented on the fact that our works stood out from the rest deep learning approaches, as they have more intuition and transparency, as well as offering a lightweight solution, which was especially appealing to some industry representatives.

Peripheral to the ICASSP venue, other non-technical activities took place, such as traditional cooking or Greek dancing lessons, and some festival nights with live music bands.

— Vasileios Magoulianitis

By |June 11th, 2023|News|Comments Off on MCL member Vasileios Magoulianitis presented in the ICASSP 2023 conference|

MCL Research on Camouflage Object Detection

Camouflage is an attempt to mask the object into a background image and to match its background. The term “camouflage” originates in the ancient practices of the animal kingdom, where animals would alter their body patterns, textures, and colors to blend in with their environment to evade predators. In military contexts, camouflage is a technique used to conceal soldiers or equipment within the background texture, making it difficult for the enemy to detect them. On the other hand, camouflaged object detection is a method used to uncover the enemy who has used camouflage to hide within the image texture.

Our research objectives are to understand the problem with more insights and develop a pipeline for Camouflaged Object Detection based on the Green Learning framework. The first main technical accomplishments are multiscale color and texture decomposition; since, in this task, the texture is much more important than the color, we decouple the information of textures and colors. The second main technical accomplishment is clustering the images dataset; we cluster similar photos with the matching color histogram together and try to learn the information. It reduces the difficulty of the classifier to separate all kinds of images, and we can focus more on images with similar colors and appearance. We cluster the dataset using HSV color space, quantize the color space into 52 bins, and use K-means clustering to derive the clusters. As shown in Figure, images with similar backgrounds are grouped together. Our experiment shows that training in each subcluster could further reduce training and validation loss.

— Max Chen

By |June 4th, 2023|News|Comments Off on MCL Research on Camouflage Object Detection|
  • Permalink Gallery

    MCL member Ganning Zhao presented in the SPIE DCS conference

MCL member Ganning Zhao presented in the SPIE DCS conference

Ganning had a trip to Orlando to participate in the SPIE DCS conference. Let’s hear what she would like to share about her work:

Ensuring the realism of computer-generated synthetic images is crucial to deep neural network (DNN) training. Due to different semantic distributions between synthetic and real-world captured datasets, there exists semantic mismatch between synthetic and refined images, which in turn results in the semantic distortion. Recently, contrastive learning (CL) has been successfully used to pull correlated patches together and push uncorrelated ones apart. In this work, we exploit semantic and structural consistency between synthetic and refined images and adopt CL to reduce the semantic distortion. Besides, we incorporate hard negative mining to improve the performance furthermore. We compare the performance of our method with several other benchmarking methods using qualitative and quantitative measures and show that our method overs the state-of-the-art performance.

— Ganning Zhao

By |May 28th, 2023|News|Comments Off on MCL member Ganning Zhao presented in the SPIE DCS conference|

MCL Genealogical Ancestry Series

Martin Ohm (May 6, 1792 in Erlangen – April 1, 1872 in Berlin) was a German mathematician and a younger brother of physicist Georg Ohm. He earned his doctorate in 1811 at Friedrich-Alexander-University, Erlangen-Nuremberg where his advisor was Karl Christian von Langsdorf. In 1817, he was appointed professor of mathematics and physics in the gymnasium at Thorn. In 1821 he moved to Berlin, and in 1839 became a full professor in the University of Berlin. He delivered courses of lectures at the academy of architecture from 1824 to 1831, and at the schools of artillery and engineering from 1833 to 1852; and he also taught in the military school from 1826 to 1849. Ohm was the first to fully develop the theory of the exponential ab when both a and b are complex numbers in 1823. The 1835 second edition of Ohm’s textbook, Die reine Elementar Mathematik was the first time that Euclid’s ‘extreme and mean ratio’ was given the name of the “golden section” (goldener Schnitt). Ohm’s notable students included Rudolf Lipschitz.

— Xiou Ge

Source: https://en.wikipedia.org/wiki/Martin_Ohm

By |May 21st, 2023|News|Comments Off on MCL Genealogical Ancestry Series|
  • Permalink Gallery

    Congratulations to MCL Members in Attending PhD Hooding Ceremony

Congratulations to MCL Members in Attending PhD Hooding Ceremony

Ten MCL members attended the Viterbi PhD hooding ceremony on Wednesday, May 10, 2023, in the Bovard Auditorium. They were Pranav Kadam, Xiou Ge, Yao Zhu, Xuejing Lei, Zohreh Azizi, Zhiruo Zhou, Yun Cheng Wang, Yifan Wang, Hongyu Fu, Hong-Shuo Chen. Congratulations to them for their accomplishments in completing their PhD program at USC!

Pranav Kadam received his Master’s degree in Electrical Engineering from USC in 2020 and Bachelor’s degree from Savitribai Phule Pune University, India in 2018. He joined Media Communications Lab in Summer 2019 guided by Prof. C.-C. Jay Kuo. His research interests include 3D Computer Vision and Machine Learning. His thesis is titled “Green Learning for 3D Point Cloud Data Processing”. He will join Tencent America.

Xiou Ge received the B.S. and M.S. degree in electrical and computer engineering from University of Illinois Urbana-Champaign, Urbana, Illinois, in 2016 and 2018 respectively. His research is on knowledge graph embedding and its applications in natural language processing. He is a recipient of USC Annenberg Graduate Fellowship (2019). His thesis is titled “Advanced Knowledge Graph Embedding Techniques: Theory and Applications”. He will join Apple’s AI/ML research team.

Yao Zhu received her Bachelor’s degree in Central South University in Changsha, Hunan, China. She joined Media Communication Lab leading by Professor Kuo in summer 2018. Her research interests include Image Synthesis, 3D construction and Computer Vision.

Xuejing Lei received her Bachelor’s degree in Automation from Xi’an Jiaotong University, China in June 2016, and received her Master’s degree in Electrical Engineering from USC in May 2018. She joined Media Communications Lab in 2017 summer. Her research interests include computer vision and deep learning.

Zohreh Azizi received the B.S. degree in Electrical Engineering from Sharif University of Technology, Iran, in 2018. She joined [...]

By |May 14th, 2023|News|Comments Off on Congratulations to MCL Members in Attending PhD Hooding Ceremony|

Congratulations to Xiou Ge for Passing His Defense

Let us hear what he has to say about his defense and an abstract of his thesis.

Knowledge graphs (KGs) find rich applications in knowledge management and discovery, recommendation systems, fraud detection, chatbots, etc. KGs are directed relational graphs. They are formed by a collection of triples in the form of (head, relation, tail), where heads and tails are called entities and represented by nodes while relations are links in KGs. Since KGs are often incomplete, one critical task in KG is to predict missing information based on available knowledge. Knowledge graph embedding (KGE) is one of the most popular approaches for link prediction and entity typing due to its effectiveness, efficiency, and explainability.

In this thesis, we propose a KGE inspired by compounding three-dimensional geometric transformation named CompoundE3D. Specifically, we apply an adapted beam search algorithm to discover the best performing variants and experiment with different ensemble strategies to achieve state-of-the-art results on four benchmarking datasets. We also analyze the properties of different operators under the CompoundE3D framework and empirically verify their advantages.

I’d like to thank Prof. Kuo for his advising throughout the last 4 years. The working style in MCL has been very effective. Without the weekly reports and meetings, it would probably take me more time to grow. I really respect Prof. Kuo’s research vision and the research direction we embark on will be appreciated in the long term. I’m also very fortunate to be with a group of very nice fellow students in the lab. Everyone is so supportive and willing to help each other. MCL is a perfect fit for students like me who may not be the most talented but are determined to work hard and do good research.

— Xiou [...]

By |May 7th, 2023|News|Comments Off on Congratulations to Xiou Ge for Passing His Defense|

MCL Research on Advanced Object Tracking Technology

Visual tracking is an important task in computer vision and has been integrated into many applications such as autonomous vehicles. Single object tracking (SOT) is the fundamental problem of visual tracking where the tracking is performed on one specific object in the testing video. The ground truth bounding box is provided in the initial frame and the tracker needs to track the object location in all later frames. There could be various challenges such as occlusion, fast motion, viewpoint change, background clutters and so on. The SOT problem has been investigated for a long time, and supervised trackers that are trained offline with large scale labeled data dominate the leader board. The model complexity and computational cost of those large deep neural networks are unaffordable for edge devices. In recent years, unsupervised learning from unlabeled data and lightweight structures start to attract the attention of researchers.

Aiming at lightweight tracking with low computational complexity, we proposed a tracker that learns from the testing sequence per se and does not require any offline training, which further extended the boundary of tracking with low level vision features with comparable performance with recent unsupervised state-of-the-art deep trackers. The number of parameters and the required flops during inference time are much smaller than those of deep trackers. Some sample results are provided in the figures above. Our tracker is able to produce flexible bounding boxes and goes back to the object after occlusion or disappearance. We hope that this work can contribute to the innovations of lightweight tracking and help with better understanding of the roles of feature representations and offline training.

–Zhiruo Zhou

By |April 30th, 2023|News|Comments Off on MCL Research on Advanced Object Tracking Technology|
  • Permalink Gallery

    Toward Interpretable, Reliable, and Sustainable AI – the 2nd AI Wave

Toward Interpretable, Reliable, and Sustainable AI – the 2nd AI Wave

Rapid advances in artificial intelligence (AI) in the last decade have been primarily attributed to the applications of deep learning (DL) technologies. DL technologies have been widely applied to speech, audio, image, video, computer graphics, 3D models, and chatbots (e.g., ChatGPTs). These advances are viewed as the first AI wave. There are concerns with the first AI wave. DL solutions are a black box (i.e., not interpretable) and vulnerable to adversarial attacks (i.e., unreliable). Besides, the high carbon footprint yielded by large DL networks is a threat to our environment (i.e., not sustainable).

Many researchers are looking for an alternative solution that is interpretable, reliable, and sustainable. This is expected to be the second AI wave. To this end, MCL members have conducted research on green learning (GL) since 2015. Since GL was inspired by DL, the two share some similarities. As time goes by, the two become more and more different. Low carbon footprints, small model sizes, low computational complexity, and mathematical transparency characterize GL. It offers energy-effective solutions in cloud centers and mobile/edge devices. It has three main modules: 1) unsupervised representation learning, 2) supervised feature learning, and 3) decision learning. GL has been successfully applied to a few applications.

The fundamental ideas of GL solution, its demonstrated examples, and its technical outlook were detailed in a recent overview paper by Kuo and Madni, “Green learning: Introduction, examples, and outlook.” Journal of Visual Communication and Image Representation (2022): 103685. We expect to see more GL solutions and applications emerging. All MCL members will be fully devoted to this field in the next decade.

By |March 19th, 2023|News|Comments Off on Toward Interpretable, Reliable, and Sustainable AI – the 2nd AI Wave|

MCL Genealogical Ancestry Series: Johann Friedrich Pfaff

Zhiruo Zhou studied Johann Friedrich Pfaff and shared her study with MCL members in the pre-seminar sharing on Mar 6th, 2023. Johann Friedrich Pfaff was born 22 December 1765 in Stuttgart, Württemberg (now Germany), and died 21 April 1825 in Halle, Saxony (now Germany). He was an influential German mathematician and well known for his work on systems of partial differential equations.

Johann Friedrich was the second son in the family and had six brothers. Although he was probably the one that gained the most fame among his siblings, his brothers were also talented in various fields. His youngest brother Johann Wilhelm Pfaff was a mathematician, and his second youngest brother Christoph Heinrich Pfaff had experiences working on chemistry, medicine, pharmacy and electricity on animals.

Johann Friedrich was born in a family with the tradition of working for the government and therefore took the school named Hohe Karlsschule to receive training for government officials’ sons when he was nine years old. He spent around eleven years there and left when he had completed studies in law which was supposed to be a fitting subject for a civil servant. Driven by the interest in mathematics and encouragement from academy, Johann Friedrich decided to move toward scientific topics and studied mathematics and physics at the University of Göttingen for two years. Then he moved to Berlin to study astronomy and wrote his first paper on astronomy there.

In 1788, Johann Friedrich accepted the election as a professor of mathematics at the University of Helmstedt and stay there until 1810. He spent a lot of efforts on teaching and increasing the number of students of mathematics. He met the student Gauss there and had a good relationship with Gauss. However, [...]

By |March 12th, 2023|News|Comments Off on MCL Genealogical Ancestry Series: Johann Friedrich Pfaff|

MCL Research on Advanced Knowledge Graph Embedding

Knowledge graph embedding has been intensively studied in recent years. TransE, RotatE, and PairRE are the most representative and effective among distance-based KG embeddings that make use of simple translation, rotation, and scaling operations respectively to represent relations between entities. These models use simple geometric transformations yet achieve relatively good performance in link prediction. However, it is unclear how to leverage the strength of these individual operations and produce a better embedding model. It turns out that we can treat each operation as a basic building block, from which we can build the most effective KG embedding scoring functions. The composition of geometric operators is a well-established tool for image manipulation. It is natural to apply the same strategy to combine translation, rotation, and scaling operators in the 3D space. Our previous work CompoundE has demonstrated that the composition of affine operations can be an effective strategy in designing KG embedding. We are interested to know whether we can achieve even better results by cascading 3D operators.

We propose to extend our previous work by including more affine operations other than only translation, rotation, and scaling in our framework, and moving from 2D transformations to 3D transformations. In addition, we propose to enhance our CompoundE by addressing two critical issues. First, although Compounding operations lead to plenty of model variants, it is still unclear how to systematically search for a scoring function that performs the best for an individual dataset. We propose to use beam search to gradually build more complex scoring functions from simple but effective variants. To speed up the model tuning during the search, we use the pre-trained entity embedding from base models as initialization. Second, although ensemble learning is a popular [...]

By |February 26th, 2023|News|Comments Off on MCL Research on Advanced Knowledge Graph Embedding|