Leveraging Multiple Features for Image Retrieval and Matching

The goal of image retrieval and matching is to find and locate object instances in images from a large-scale image database. While visual features are abundant, how to combine them to improve performance by individual features remains a challenging task. In this work, we focus on leveraging multiple features for accurate and efficient image retrieval and matching. We first propose two graph-based approaches to rerank initially retrieved images for generic image retrieval. In the graph, vertices are images while edges are similarities between image pairs. Our first approach employs a mixture Markov model based on a random walk model on multiple graphs to fuse graphs. We introduce a probabilistic model to compute the importance of each feature for graph fusion under a naive Bayesian formulation, which requires statistics of similarities from a manually labeled dataset containing irrelevant images. To reduce human labeling, we further propose a fully unsupervised reranking algorithm based on a submodular objective function that can be efficiently optimized by greedy algorithm. By maximizing an information gain term over the graph, our submodular function favors a subset of database images that are similar to query images and resemble each other. The function also exploits the rank relationships of images from multiple ranked lists obtained by different features. We then study a more well-defined application, person re-identification, where the database contains labeled images of human bodies captured by multiple cameras. Re-identifications from multiple cameras are regarded as related tasks to exploit shared information. We apply a novel multi-task learning algorithm using both low level features and attributes. A low rank attribute embedding is joint learned within the multi-task learning formulation to embed original binary attributes to a continuous attribute space, where incorrect and incomplete attributes are rectified and recovered. To locate objects in images, we design an object detector based on object proposals and deep convolutional neural networks (CNN) in view of the emergence of deep networks. We improve a Fast RCNN framework and investigate two new strategies to detect objects accurately and efficiently: scale-dependent pooling (SDP) and cascaded rejection classifiers (CRC). The SDP improves detection accuracy by exploiting appropriate convolutional features depending on the scale of input object proposals. The CRC effectively utilizes convolutional features and greatly eliminates negative proposals in a cascaded manner, while maintaining a high recall for true objects. The two strategies together improve the detection accuracy and reduce the computational cost.

A von Bertalanffy Based Model for the Estimation of Oyster (Crassostrea virginica) Growth on Restored Oyster Reefs in Chesapeake Bay

A model to estimate the mean monthly growth of Crassostrea virginica oysters in Chesapeake Bay was developed. This model is based on the classic von Bertalanffy growth function, however the growth constant is changed every monthly timestep in response to short term changes in temperature and salinity. Using a dynamically varying growth constant allows the model to capture seasonal oscillations in growth, and growth responses to changing environmental conditions that previous applications of the von Bertalanffy model do not capture. This model is further expanded to include an estimation of Perkinsus marinus impacts on growth rates as well as estimations of ecosystem services provided by a restored oyster bar over time. The model was validated by comparing growth estimates from the model to oyster shell height observations from a variety of restoration sites in the upper Chesapeake Bay. Without using the P. marinus impact on growth, the model consistently overestimates mean oyster growth. However, when P. marinus effects are included in the model, the model estimates match the observed mean shell height closely for at least the first 3 years of growth. The estimates of ecosystem services suggested by this model imply that even with high levels of mortality on an oyster reef, the ecosystem services provided by that reef can still be maintained by growth for several years. Because larger oyster filter more water than smaller ones, larger oysters contribute more to the filtration and nutrient removal ecosystem services of the reef. Therefore a reef with an abundance of larger oysters will provide better filtration and nutrient removal. This implies that if an oyster restoration project is trying to improve water quality through oyster filtration, it is important to maintain the larger older oysters on the reef.