This paper presents a new 3D time-space detector for small ships in single look complex (SLC) synthetic aperture radar (SAR) imagery, optimized for small targets around 5-15 m long that are unfocused due to target motion induced by ocean surface waves.
Multi-modal data analytics publications
Deep Adaptive Semantic Logic (DASL): Compiling Declarative Knowledge into Deep Neural Networks
We introduce Deep Adaptive Semantic Logic (DASL), a novel framework for automating the generation of deep neural networks that incorporates user-provided formal knowledge to improve learning from data. We provide formal semantics that demonstrate that our knowledge representation captures all of first order logic and that finite sampling from infinite domains converges to correct truth values. DASL’s representation improves on prior neural-symbolic work by avoiding vanishing gradients, allowing deeper logical structure, and enabling richer interactions between the knowledge and learning components. We illustrate DASL through a toy problem in which we add structure to an image classification problem and demonstrate that knowledge of that structure reduces data requirements by a factor of 1000 . We then evaluate DASL on a visual relationship detection task and demonstrate that the addition of commonsense knowledge improves performance by 10.7 % in a data scarce setting.
Stacked Spatio-Temporal Graph Convolutional Networks for Action Segmentation
Abstract We propose novel Stacked Spatio-Temporal Graph Convolutional Networks (Stacked-STGCN) for action segmentation, i.e., predicting and localizing a sequence of actions over long videos. We extend the Spatio-Temporal Graph Convolutional Network (STGCN) originally proposed for skeleton-based action recognition to enable nodes with different characteristics (e.g., scene, actor, object, action, etc.), feature descriptors with varied lengths, […]
Aesop: A Visual Storytelling Platform for Conversational AI and Commonsense Grounding
We believe that the future of Artificial Intelligence (AI) will be a mixed-initiative collaboration between humans and AI as equals.
Efficient Fine-Grained Classification and Part Localization Using One Compact Network
Abstract Fine-grained classification of objects such as vehicles, natural objects and other classes is an important problem in visual recognition. It is a challenging task because small and localized differences between similar looking objects indicate the specific fine-grained label. At the same time, accurate classification needs to discount spurious changes in appearance caused by occlusions, […]
Analyzing hyperspectral images into multiple subspaces using Gaussian mixture models
I argue that the spectra in a hyperspectral datacube will usually lie in several low-dimensional subspaces, and that these subspaces are more easily estimated from the data than the endmembers. I present an algorithm for finding the subspaces. The algorithm fits the data with a Gaussian mixture model, in which the means and covariance matrices are parameterized in terms of the subspaces. The locations of materials can be inferred from the fit of library spectra to the subspaces. The algorithm can be modified to perform material detection. This has better performance than standard algorithms such as ACE, and runs in real time.
Pattern of life analysis for diverse data types
SRI has developed a system to automatically analyze the Pattern of Life (PoL) of ports, routes and vessels from a large collection of AIS data. The PoL of these entities are characterized by a set of intuitive and easy to query semantic attributes. The prototype system provides an interface to ingest other types of information such as WAAS (Wide Area Aerial Surveillance) and GDELT (Global Database of Events, Language, and Tone) to augment knowledge of the Area of Operations. It can interact with users by answering questions and simulating what-if scenarios to keep human in the processing loop.
Sampling Exactly from the Normal Distribution
An algorithm for sampling exactly from the normal distribution is given. The algorithm reads some number of uniformly distributed random digits in a given base and generates an initial portion of the representation of a normal deviate in the same base. Thereafter, uniform random digits are copied directly into the representation of the normal deviate. Thus, in contrast to existing methods, it is possible to generate normal deviates exactly rounded to any precision with a mean cost that scales linearly in the precision. The method performs no extended precision arithmetic, calls no transcendental functions, and uses no floating point arithmetic whatsoever; it uses only simple integer operations. It can easily be adapted to sample exactly from the discrete normal distribution whose parameters are rational numbers.
Re-Ranking by Multi-Feature Fusion with Diffusion for Image Retrieval
We present a re-ranking algorithm for image retrieval by fusing multi-feature information. We utilize pairwise similarity scores between images to exploit the underlying relationships among images. The initial ranked list for a query from each feature is represented as an undirected graph, where edge strength comes from feature-specific image similarity. Graphs from multiple features are combined by a mixture Markov model. In addition, we utilize a probabilistic model based on the statistics of similarity scores of similar and dissimilar image pairs to determine the weight for each graph. The weight for a feature is queryspecific, where the ranked lists of different queries receive different weights. Our approach for calculating weights is data-driven and does not require any learning. A diffusion process is then applied to the fused graph to reduce noise and achieve better retrieval performance. Experiments demonstrate that our approach significantly improves performance over baseline methods and outperforms many state-of-the-art retrieval methods.