Weighted boxes fusion: Ensembling boxes from different object detection models

Edge Boxes: Locating Object Proposals from Edges

The use of object proposals is an effective recent approach for increasing the computational efficiency of object detection. We propose a novel method for generating object bounding box proposals using edges. Edges provide a sparse yet informative representation of an image. Our main observation is that the number of contours that are wholly contained in a bounding box is indicative of the like...

Video Object Detection and Segmentation Based on Proposal Boxes

Understading Black Boxes: Knowledge Induction From Models

Due to regurations and laws prohibiting uses of private data on customers and their transactions in customer data base, most customer data sets are not easily accessable even in the same organizations. A solutio for this reguatory problems can be providing statistical summary of the data or models induced from the dat, instead of providing raw data sets. The models, however, have limited inform...

Boxes: A Symbolic Abstract Domain of Boxes

Numeric abstract domains are widely used in program analyses. The simplest numeric domains over-approximate disjunction by an imprecise join, typically yielding path-insensitive analyses. This problem is addressed by domain refinements, such as finite powersets, which provide exact disjunction. However, developing correct and efficient disjunctive refinement is challenging. First, there must be...

Common Developments of Three Different Orthogonal Boxes

We investigate common developments that can fold into plural incongruent orthogonal boxes. It was shown that there are infinitely many orthogonal polygons that fold into two incongruent orthogonal boxes in 2008. In 2011, it was shown that there exists an orthogonal polygon that folds into three boxes of size 1 × 1 × 5, 1 × 2 × 3, and 0×1×11. It remained open whether there exists an orthogonal p...