Fine-grained image recognition is a challenging computer vision problem, due to the small inter-class variations caused by highly similar subordinate categories, and the large intra-class variations in poses, scales and rotations. In this paper, we propose a novel end-to-end Mask-CNN model without the fully connected layers for fine-grained recognition. Based on the part annotations of fine-gra...

Semantic part localization can facilitate fine-grained categorization by explicitly isolating subtle appearance differences associated with specific object parts. Methods for pose-normalized representations have been proposed, but generally presume bounding box annotations at test time due to the difficulty of object detection. We propose a model for fine-grained categorization that overcomes t...

Electron micrographs of directcarbon replicas having magnifications of 5400 to 23, 500 times reveal in detail the grain size, shape, and packing of the particles comprising very fine-grained limestones. At these magnifications, however, only portions of a few grains in mediumor coarsegrained limestones can be observed. Grain contacts are mostly curvilinear in very fine-grained limestones but ar...

Existing approaches for fine-grained access control for RDF data suffer from high overhead, making them ill-suited for mobile devices. This makes it difficult to develop mobile applications that manage personal RDF data in a privacy preserving manner. In this paper we propose a new approach to realise fine-grained access control for mobile devices. We show how fine-grained privacy settings for ...

This paper summarizes the annotation of fine-grained entailment relationships in the context of student answers to science assessment questions. We annotated a corpus of 15,357 answer pairs with 145,911 fine-grained entailment relationships. We provide the rationale for such fine-grained analysis and discuss its perceived benefits to an Intelligent Tutoring System. The corpus also has potential...

State of the art semi-supervised entity set expansion algorithms produce noisy results, which need to be refined manually. Sets expanded for intended fine-grained concepts are especially noisy because these concepts are not well represented by the limited number of seeds. Such sets are usually incorrectly expanded to contain elements of a more general concept. We show that fine-grained control ...

The island of Syros consists largely of possibly-repeated sequences of glaucophane-bearing calcareous schists, mafic schists, dolomite marbles, and calcite marbles containing abundant aragonite pseudomorphs (Dixon, 1969; Hecht, 1984). Several discrete, fault-bounded packages of blueschist/eclogite-facies mafic rocks with minor serpentinite are also found on the island. Although the mafic rocks ...

In many QA systems, fine-grained named entities are extracted by coarse-grained named entity recognizer and fine-grained named entity dictionary. In this paper, we describe a fine-grained Named Entity Recognition using Conditional Random Fields (CRFs) for question answering. We used CRFs to detect boundary of named entities and Maximum Entropy (ME) to classify named entity classes. Using the pr...

Global sparse analysis framework aims to design precise, sound, yet scalable static analyzers systematically. The framework breaks the tradeoff between general-purpose but coarse-grained sparse analyses and fine-grained but limited-purpose sparse analyses by providing a general-purpose fine-grained sparse analysis framework. However, the formal aspect of the framework has not been presented rig...

Current lexical semantic theories provide representations at a coarse grained level. In this paper, I will provide motivations for a fine grained representation for verbs and. nouns. An initial case study is done to serve as evidence that a more detailed representation is needed for tasks that require high accuracy rates, such as machine translation. An automatic approach to gather fine grained...