No abstract

This paper presents a simple yet effective supervised deep hash approach that constructs binary hash codes from labeled data for large-scale image search. We assume that the semantic labels are governed by several latent attributes with each attribute on or off, and classification relies on these attributes. Based on this assumption, our approach, dubbed supervised semantics-preserving deep hashing (SSDH), constructs hash functions as a latent layer in a deep network and the binary codes are learned by minimizing an objective function defined over classification error and other desirable hash codes properties. With this design, SSDH has a nice characteristic that classification and retrieval are unified in a single learning model. Moreover, SSDH performs joint learning of image representations, hash codes, and classification in a point-wised manner, and thus is scalable to large-scale datasets. SSDH is simple and can be realized by a slight enhancement of an existing deep architecture for classification; yet it is effective and outperforms other hashing approaches on several benchmarks and large datasets. Compared with state-of-the-art approaches, SSDH achieves higher retrieval accuracy, while the classification performance is not sacrificed.

No abstract

Sperm quality assessment plays an essential role in human fertility and animal breeding. Manual analysis is time-consuming and subject to intra-and inter-observer variability. To automate the analysis process, as well as to offer a means of statistical analysis that may not be achieved by visual inspection, we present a computational framework that tracks the heads and traces the tails for analyzing sperm motility, one of the most important attributes in semen quality evaluation. Our framework consists of 3 modules: head detection, head tracking, and flagellum tracing. The head detection module detects the sperm heads from the image data, and the detected heads are the inputs to the head tracking module for obtaining the head trajectories. Finally, a flagellum tracing algorithm is proposed to obtain the flagellar beat patterns. Our framework aims at providing both the head trajectories and the flagellar beat patterns for quantitatively assessing sperm motility. This distinguishes our work from other existing methods that analyze sperm motility based merely on the head trajectories. We validate our framework using two confocal microscopy image sequences of ram semen samples that were imaged at two different conditions, at which the sperms behave differently. The results show the effectiveness of our framework.

What is available to developmental programs in autonomous mental development, and what should be learned at the very early stages of mental development? Our observation is that sensory and motor primitives are the most basic components present at the beginning, and what developmental agents need to learn from these resources is what their internal sensory states stand for. In this paper, we investigate the question in the context of a simple biologically motivated visuomotor agent. We observe and acknowledge, as many other researchers do, that action plays a key role in providing content to the sensory state. We propose a simple, yet powerful learning criterion, that of invariance, where invariance simply means that the internal state does not change over time. We show that after reinforcement learning based on the invariance criterion, the property of action sequence based on an internal sensory state accurately reflects the property of the stimulus that triggered that internal state. That way, the meaning of the internal sensory state can be firmly grounded on the property of that particular action sequence. We expect the framing of the problem and the proposed solution presented in this paper to help shed new light on autonomous understanding in developmental agents such as humanoid robots.