Content-Aware Focal Plane Selection and Proposals for Object Tracking on Plenoptic Image Sequences.

Object tracking is a fundamental problem in computer vision since it is required in many practical applications including video-based surveillance and autonomous vehicles. One of the most challenging scenarios in the problem is when the target object is partially or even fully occluded by other objects. In such cases, most of existing trackers can fail in their task while the object is invisible. Recently, a few techniques have been proposed to tackle the occlusion problem by performing the tracking on plenoptic image sequences. Although they have shown promising results based on the refocusing capability of plenoptic images, there is still room for improvement. In this paper, we propose a novel focus index selection algorithm to identify an optimal focal plane where the tracking should be performed. To determine an optimal focus index, we use a focus measure to find maximally focused plane and a visual similarity to capture the plane where the target object is visible, and its appearance is distinguishably clear. We further use the selected focus index to generate proposals. Since the optimal focus index allows us to estimate the distance between the camera and the target object, we can more accurately guess the scale changes of the object in the image plane. Our proposal algorithm also takes the trajectory of the target object into account. We extensively evaluate our proposed techniques on three plenoptic image sequences by comparing them against the prior tracking methods specialized to the plenoptic image sequences. In experiments, our method provides higher accuracy and robustness over the prior art, and those results confirm that the merits of our proposed algorithms.

Frequency-Based Motion Representation for Video Generative Adversarial Networks.

Videos contain motions of various speeds. For example, the motions of one's head and mouth differ in terms of speed - the head being relatively stable and the mouth moving rapidly as one speaks. Despite its diverse nature, previous video GANs generate video based on a single unified motion representation without considering the aspect of speed. In this paper, we propose a frequency-based motion representation for video GANs to realize the concept of speed in video generation process. In detail, we represent motions as continuous sinusoidal signals of various frequencies by introducing a coordinate-based motion generator. We show, in that case, frequency is highly related to the speed of motion. Based on this observation, we present frequency-aware weight modulation that enables manipulation of motions within a specific range of speed, which could not be achieved with the previous techniques. Extensive experiments validate that the proposed method outperforms state-of-the-art video GANs in terms of generation quality by its capability to model various speed of motions. Furthermore, we also show that our temporally continuous representation enables to further synthesize intermediate and future frames of generated videos.

Genetic variants of the FABP4 gene are associated with marbling scores and meat quality grades in Hanwoo (Korean cattle).

This study was aimed to search new genetic variants in the bovine FABP4 gene as molecular markers for meat quality and carcass traits. PCR-RFLP analysis revealed that three SNPs located at nucleotide positions g.2834C>G, g.3533T>A, and g.3691G>A were identified based on a GenBank accession number (NC_007312.4). Sequence analysis revealed that SNPs were located in intron 1 (g.2834C>G) and 2 (g.3533T>A), and an exon 3 (g.3691G>A), showing allele frequencies as 0.592, 0.579, and 0.789, respectively. Genetic variabilities of heterozygosity (He) and polymorphic information contents (PIC) were estimated for g.2834C>G (0.608 and 0.531), g.3533T>A (0.615 and 0.539), and g.3691G>A (0.498 and 0.401) loci, respectively. A SNP located in the exon 3 of FABP4 was characterized and associated with desirable increases of MS (marbling scores) and MG (meat quality grades) in Hanwoo. The statistical analysis revealed that additive effects by GG genotypes in g.3691G>A SNP were significantly greater than AA genotypes in MS and MG traits. These findings suggest that the FABP4g.3691G>A SNP will be a useful candidate locus to maximize economic benefits for cattle populations.

Distance Encoded Product Quantization for Approximate K-Nearest Neighbor Search in High-Dimensional Space.

Approximate K-nearest neighbor search is a fundamental problem in computer science. The problem is especially important for high-dimensional and large-scale data. Recently, many techniques encoding high-dimensional data to compact codes have been proposed. The product quantization and its variations that encode the cluster index in each subspace have been shown to provide impressive accuracy. In this paper, we explore a simple question: is it best to use all the bit-budget for encoding a cluster index? We have found that as data points are located farther away from the cluster centers, the error of estimated distance becomes larger. To address this issue, we propose a novel compact code representation that encodes both the cluster index and quantized distance between a point and its cluster center in each subspace by distributing the bit-budget. We also propose two distance estimators tailored to our representation. We further extend our method to encode global residual distances in the original space. We have evaluated our proposed methods on benchmarks consisting of GIST, VLAD, and CNN features. Our extensive experiments show that the proposed methods significantly and consistently improve the search accuracy over other tested techniques. This result is achieved mainly because our methods accurately estimate distances.

Spherical hashing: binary code embedding with hyperspheres.

Many binary code embedding schemes have been actively studied recently, since they can provide efficient similarity search, and compact data representations suitable for handling large scale image databases. Existing binary code embedding techniques encode high-dimensional data by using hyperplane-based hashing functions. In this paper we propose a novel hypersphere-based hashing function, spherical hashing, to map more spatially coherent data points into a binary code compared to hyperplane-based hashing functions. We also propose a new binary code distance function, spherical Hamming distance, tailored for our hypersphere-based binary coding scheme, and design an efficient iterative optimization process to achieve both balanced partitioning for each hash function and independence between hashing functions. Furthermore, we generalize spherical hashing to support various similarity measures defined by kernel functions. Our extensive experiments show that our spherical hashing technique significantly outperforms state-of-the-art techniques based on hyperplanes across various benchmarks with sizes ranging from one to 75 million of GIST, BoW and VLAD descriptors. The performance gains are consistent and large, up to 100 percent improvements over the second best method among tested methods. These results confirm the unique merits of using hyperspheres to encode proximity regions in high-dimensional spaces. Finally, our method is intuitive and easy to implement.