We describe a method for implementing the evaluation and training of decision trees and forests e... more We describe a method for implementing the evaluation and training of decision trees and forests entirely on a GPU, and show how this method can be used in the context of object recognition. Our strategy for evaluation involves mapping the data structure describing a decision forest to a 2D texture array. We navigate through the forest for each point of the input data in parallel using an efficient, non-branching pixel shader. For training, we compute the responses of the training data to a set of candidate features, and scatter the responses into a suitable histogram using a vertex shader. The histograms thus computed can be used in conjunction with a broad range of tree learning algorithms. We demonstrate results for object recognition which are identical to those obtained on a CPU, obtained in about 1% of the time. To our knowledge, this is the first time a method has been proposed which is capable of evaluating or training decision trees on a GPU. Our method leverages the full parallelism of the GPU. Although we use features common to computer vision to demonstrate object recognition, our framework can accommodate other kinds of features for more general utility within computer science.
A real-life requirement motivated this case study of secure covert communication. An independentl... more A real-life requirement motivated this case study of secure covert communication. An independently researched process is described in detail with an emphasis on implementation issues regarding digital images. A scheme using stego keys to create pseudorandom sample sequences is developed. Issues relating to using digital signals for steganography are explored. The terms modified remainder and unmodified remainder are defined. Possible attacks are considered in detail from passive wardens and methods of defeating such attacks are suggested. Software implementing the new ideas is introduced, which has been successfully developed, deployed and used for several years without detection.
We propose a new method to quickly and accurately predict 3D positions of body joints from a sing... more We propose a new method to quickly and accurately predict 3D positions of body joints from a single depth image, using no temporal information. We take an object recognition approach, designing an intermediate body parts representation that maps the difficult pose estimation problem into a simpler per-pixel classification problem. Our large and highly varied training dataset allows the classifier to estimate body parts invariant to pose, body shape, clothing, etc. Finally we generate confidence-scored 3D proposals of several body joints by reprojecting the classification result and finding local modes. The system runs at 200 frames per second on consumer hardware. Our evaluation shows high accuracy on both synthetic and real test sets, and investigates the effect of several training parameters. We achieve state of the art accuracy in our comparison with related work and demonstrate improved generalization over exact whole-skeleton nearest neighbor matching.
We describe a method for implementing the evaluation and training of decision trees and forests e... more We describe a method for implementing the evaluation and training of decision trees and forests entirely on a GPU, and show how this method can be used in the context of object recognition. Our strategy for evaluation involves mapping the data structure describing a decision forest to a 2D texture array. We navigate through the forest for each point of the input data in parallel using an efficient, non-branching pixel shader. For training, we compute the responses of the training data to a set of candidate features, and scatter the responses into a suitable histogram using a vertex shader. The histograms thus computed can be used in conjunction with a broad range of tree learning algorithms. We demonstrate results for object recognition which are identical to those obtained on a CPU, obtained in about 1% of the time. To our knowledge, this is the first time a method has been proposed which is capable of evaluating or training decision trees on a GPU. Our method leverages the full parallelism of the GPU. Although we use features common to computer vision to demonstrate object recognition, our framework can accommodate other kinds of features for more general utility within computer science.
A real-life requirement motivated this case study of secure covert communication. An independentl... more A real-life requirement motivated this case study of secure covert communication. An independently researched process is described in detail with an emphasis on implementation issues regarding digital images. A scheme using stego keys to create pseudorandom sample sequences is developed. Issues relating to using digital signals for steganography are explored. The terms modified remainder and unmodified remainder are defined. Possible attacks are considered in detail from passive wardens and methods of defeating such attacks are suggested. Software implementing the new ideas is introduced, which has been successfully developed, deployed and used for several years without detection.
We propose a new method to quickly and accurately predict 3D positions of body joints from a sing... more We propose a new method to quickly and accurately predict 3D positions of body joints from a single depth image, using no temporal information. We take an object recognition approach, designing an intermediate body parts representation that maps the difficult pose estimation problem into a simpler per-pixel classification problem. Our large and highly varied training dataset allows the classifier to estimate body parts invariant to pose, body shape, clothing, etc. Finally we generate confidence-scored 3D proposals of several body joints by reprojecting the classification result and finding local modes. The system runs at 200 frames per second on consumer hardware. Our evaluation shows high accuracy on both synthetic and real test sets, and investigates the effect of several training parameters. We achieve state of the art accuracy in our comparison with related work and demonstrate improved generalization over exact whole-skeleton nearest neighbor matching.
Uploads
Papers by Toby Sharp