Monthly Archives: January 2017

Congratulations to Hao Xu for passing his defense on January 23, 2017. His Ph.D. thesis is entitled “Understanding Deep Learning from Its System Architectures, Feature Representations to Applications”.

Abstract of thesis:

Deep learning plays key roles in various aspects of the modern computer vision research. Our research focused on analyzing, adopting, and developing better CNN architectures which outperform the previous methods. To begin with, a car detection method using deformable part models consisting of composite feature sets (DPM/CF) is proposed. It recognizes cars of various types and from multiple viewing angles. The DPM/CF system consists of two stages. In the first stage, an HOG template is used to detect the bounding box of the entire car of a certain type and viewed from a certain angle (called a t/a pair), which yields a region of interest (ROI). In the second stage, we detect each salient part in a given t/a-specific ROI using either the HOG or the CNN feature. An optimization procedure based on latent logistic regression is adopted to choose the most discriminative location/size and the most suitable feature set for each part automatically. It is observed that the DPM/CF detector can strike a balance between detection performance and training complexity, through selecting the capable and simple feature from the composite feature set. Extensive experimental results are given to demonstrate the superior performance of the proposed DPM/CF method.

The CNN features used in DPM/CF demonstrate strong performance in detecting objects from images. To analyze the strength and weakness of the CNN feature representation, two quantitative metrics are proposed for the automatic evaluation of trained features at different convolution layers. The Gaussian confusion measure (GCM) is used to identify the discriminative ability of an individual feature, while [...]

Congratulations to Qin Huang for passing his Qualifying Exam on January 18, 2016. The title of his Ph.D. thesis proposal is “Machine Learning Techniques for Perceptual Quality Enhancement and Semantic Image Segmentation”. His qualifying exam committee consisted of Jay Kuo (Chair), Antonio Ortega, Justin Haldar, Sandy Sawchuk and Cyrus Shahabi (Outside Member).

Abstract of thesis proposal:

Researches on image processing and computer vision problems can be generally divided into two major steps: extracting powerful feature representations and designing efficient decision system. Traditional methods rely on hand-craft features, as well as pre-defined thresholds to generate a necessary condition required for the desired target. A more robust system could be designed taking advantage of machine learning techniques if enough training samples are provided. Thanks to the development of big data, millions of image and video contents are now available for training. To better utilize the information in the training, convolutional neural network based deep learning systems become popular in recent years. Specifically, the CNN based methods demonstrate better ability to acquire powerful feature representations in a simultaneous way. However, CNN based training has a high requirement of hardware and subtle process design. And therefore it should be carefully explored in order to obtain desired results.

In this proposal, we contribute to three works that gradually develop from the traditional method to deep learning based method. Based on the applications, the works can be divided into two major categories: perceptual quality enhancement and semantic image segmentation. In the first part, we focus on enhancing the quality of images and videos by considering related perceptual properties of human visual system. To begin with, we deal with a type of compression artifacts referred to as “false contour”. We then focus on the visual experience [...]

Congratulations to Eddy Wu for passing his Qualifying Exam on January 13, 2016. The title of his Ph.D. thesis proposal is “Deep Learning Techniques for Supervised and Semi-Supervised Pedestrian Detection”. His qualifying exam committee consisted of Jay Kuo (Chair), Sandy Sawchuk, Richard Leahy, Justin Haldar and Aiichiro Nakano (Outside Member).

Abstract of thesis proposal:

With the emergence of autonomous driving and the advanced driver assistance system (ADAS), the importance of pedestrian detection has increased significantly. A lot of research work has been conducted to tackle this problem with the availability of large-scale datasets. Methods based on the convolutional neural network (CNN) technology have achieved great success in pedestrian detection in recent years, which offers a giant step to the solution of this problem.  Although the performance of CNN-based solutions reaches a significantly higher level than traditional methods, it is still far from perfection. Further advancement in this field is still demanded. In this proposal, we conducted two research topics along this direction.

In the first topic, a boosted convolutional neural network (BCNN) system is proposed to enhance the pedestrian detection performance. Being inspired by the classic boosting idea, we develop a weighted loss function that emphasizes challenging samples in training a convolutional neural network (CNN). Two types of samples are considered challenging:

1) samples with detection scores falling in the decision boundary, and

2) temporally associated samples with inconsistent scores. A weighting scheme is designed for each of them. Finally, we train a boosted fusion layer to benefit from the integration of these two weighting schemes. We use the Fast-RCNN as the baseline and test the corresponding BCNN on the Caltech pedestrian dataset in the experiment and observe a significant performance gain of the BCNN over its baseline.

Data-driven pedestrian detection methods demand [...]

The MCL joined a collaborative project to build a large-scale subjective video quality database. The database was proposed to boost a major breakthrough in video coding and processing. It consists of 220 5-second sequences in four resolutions (i.e., 1920×1080, 1280×720, 960×540 and 640×360). For each of the 880 video clips, we encoded it using the H.264 codec and conducted a large-scale subjective test on the perceptual quality. The dataset is called the “VideoSet”, which is an acronym for “Video Subject Evaluation Test (SET)”. The database are available to the public in the IEEE DataPort.

IEEE DataPort is a valuable and universally accessible repository of datasets serving the growing data needs in both research and industry. The repository is designed to accept all types of datasets, including Big Data datasets up to 2TB, and it provides both downloading capabilities and access to Cloud services to enable data analysis in the Cloud.

We appreciate the help from Dr. K. J. Ray Liu and Melissa Handa in hosting the VidoeSet database.