This paper proposes a framework for automatically annotating the keypoints of a human body in images for learning 2D pose estimation models. Ground-truth annotations for supervised learning are difficult and cumbersome in most machine vision tasks. While considerable contributions in the community provide us a huge number of pose-annotated images, all of them mainly focus on people wearing common clothes, which are relatively easy to annotate the body keypoints. This paper, on the other hand, focuses on annotating people wearing loose-fitting clothes (e.g., Japanese Kimono) that occlude many body keypoints. In order to automatically and correctly annotate these people, we divert the 3D coordinates of the keypoints observed without loose-fitting clothes, which can be captured by a motion capture system (MoCap). These 3D keypoints are projected to an image where the body pose under loose-fitting clothes is similar to the one captured by the MoCap. Pose similarity between bodies with and without loose-fitting clothes is evaluated with 3D geometric configurations of MoCap markers that are visible even with loose-fitting clothes (e.g., markers on the head, wrists, and ankles). Experimental results validate the effectiveness of our proposed framework for human pose estimation.

We describe FMCW reflectometry for characterization of long optical fibers by using an external-cavity laser diode as a light source. Since the optical path difference between the reference beam and the reflected beam from the optical fiber under test is much longer than the coherence length of the light source, the reference and the reflected beams are phase-decorrelated. As a result, the beat spectrum between the reference and the reflected beams is measured. In the phase-decorrelated FMCW reflectomety, the spatial resolution is enhanced by narrowing the spectral linewidth of the light source and increasing the repetition frequency of the optical frequency sweep as well as increasing the chirping range of the optical frequency sweep. In the experiments, an external-cavity DFB laser is used as a narrow linewidth light source, and the optical frequency is swept by minute modulation of the external cavity length. Long single mode optical fibers are characterized, and the maximum measurement range of 80 km is achieved, and the spatial resolutions of 46 m, 100 m and 2 km are achieved at 5 km, 11 km and 80 km distant, respectively. The Rayleigh backscattering is clearly measured and the propagation loss of optical fiber is also measured. The optical gain of an erbium-doped optical fiber amplifier (EDFA) is also estimated from the change in the Rayleigh backscattering level in the optical fiber followed after the EDFA.