Engineering Commons^™

Deep Convolutional Neural Networks For Estimating Lens Distortion Parameters, Sebastian Lutz, Mark Davey, Aljosa Smolic

Session 2: Deep Learning for Computer Vision

In this paper we present a convolutional neural network (CNN) to predict multiple lens distortion parameters from a single input image. Unlike other methods, our network is suitable to create high resolution output as it directly estimates the parameters from the image which then can be used to rectify even very high resolution input images. As our method it is fully automatic, it is suitable for both casual creatives and professional artists. Our results show that our network accurately predicts the lens distortion parameters of high resolution images and corrects the distortions satisfactory.

No Room For Squares: Using Bitmap Masks To Improve Pedestrian Detection Using Cnns., Adam Warde, Hamza Yous, David Gregg, David Moloney

Session 3: Deep Learning for Computer Vision

.

In this paper we investigate a method to reduce the number of computations and associated activations in Convolutional Neural Networks (CNN) by using bitmaps. The bitmaps are used to mask the input images to the network that fall within a rectangular window but do not fall within the boundaries of the objects the network is being trained upon. The mask has the effect of rendering the operations on these portions of the training images trivial. The thesis is that applying this approach to CNNs will not degrade accuracy while at the same time reducing the computational workload and reducing …

Comparing Data Augmentation Strategies For Deep Image Classification, Sarah O'Gara, Kevin Mcguinness

Session 2: Deep Learning for Computer Vision

Currently deep learning requires large volumes of training data to fit accurate models. In practice, however, there is often insufficient training data available and augmentation is used to expand the dataset. Historically, only simple forms of augmentation, such as cropping and horizontal flips, were used. More complex augmentation methods have recently been developed, but it is still unclear which techniques are most effective, and at what stage of the learning process they should be introduced. This paper investigates data augmentation strategies for image classification, including the effectiveness of different forms of augmentation, dependency on the number of training examples, and …

Performance Comparison Of Hybrid Cnn-Svm And Cnn-Xgboost Models In Concrete Crack Detection, Sahana Thiyagarajan

Dissertations

Detection of cracks mainly has been a sort of essential step in visual inspection involved in construction engineering as it is the commonly used building material and cracks in them is an early sign of de-basement. It is hard to find cracks by a visual check for the massive structures. So, the development of crack detecting systems generally has been a critical issue. The utilization of contextual image processing in crack detection is constrained, as image data usually taken under real-world situations vary widely and also includes the complex modelling of cracks and the extraction of handcrafted features. Therefore the …

Investigation Into The Perceptually Informed Data For Environmental Sound Recognition, Chenglin Kang

Dissertations

Environmental sound is rich source of information that can be used to infer contexts. With the rise in ubiquitous computing, the desire of environmental sound recognition is rapidly growing. Primarily, the research aims to recognize the environmental sound using the perceptually informed data. The initial study is concentrated on understanding the current state-of-the-art techniques in environmental sound recognition. Then those researches are evaluated by a critical review of the literature. This study extracts three sets of features: Mel Frequency Cepstral Coefficients, Mel-spectrogram and sound texture statistics. Two kinds machine learning algorithms are cooperated with appropriate sound features. The models are …