1 In this paper we tackle the problem of vehicle reidentification in a camera network utilizing triplet embeddings. Re-identification is the problem of matching appearances of objects across different cameras. With the proliferation of surveillance cameras enabling smart and safer cities, there is an ever-increasing need to re-identify vehicles across cameras. Typical challenges arising in smart city scenarios include variations of viewpoints, illumination and self occlusions. Most successful approaches for reidentification involve (deep) learning an embedding space such that the vehicles of same identities are projected closer to one another, compared to the vehicles representing different identities. Popular loss functions for learning an embedding (space) include contrastive or triplet loss. In this paper we provide an extensive evaluation of these losses applied to vehicle re-identification and demonstrate that using the best practices for learning embeddings outperform most of the previous approaches proposed in the vehicle reidentification literature. Compared to most existing stateof-the-art approaches, our approach is simpler and more straightforward for training utilizing only identity-level annotations, along with one of the smallest published embedding dimensions for efficient inference. Furthermore in this work we introduce a formal evaluation of a triplet sampling variant (batch sample) into the re-identification literature.

A scanning laser Doppler technique based on Chebyshev demodulation has been developed for the rapid measurement of spatially distributed velocity profiles. Scan frequencies up to 100 Hz can be used over scan lengths up to 270 mm. The Doppler signals are processed in the conventional manner using a frequency counter. The analog velocity output from the counter is post-processed to obtain the velocity profile. The Chebyshev demodulation post-processing technique for processing the velocity signals from solid surfaces has been introduced. The data processing technique directly yields the spatial velocity distribution in approximate functional form through frequency domain analysis of the scanning LDV velocity output. Results from a rotating disk setup are presented to illustrate the concept.

No abstract

The static energy absorption behavior of graphite epoxy composite corrugated (sine wave) webs loaded in axial compression is reported in this paper. Tests have been conducted to study the effects of various geometric parameters of the web specimen including width and gross thickness. The importance of the failure initiator and its effect on energy absorption are described along with other observed energy absorption trends. Comparisons are made with published tube specimen behavior where appropriate. The existence of a stability boundary within which efficient crushing occurs is shown in the case of gross thickness variation of the web.

The discrete wavelet transform has recently emerged as a powerful technique for decomposing images into various multi-resolution approximations. Multiresolution decomposition schemes have proven to be very effective for high-quality, low bit-rate image coding. In this work, we investigate the use of entropyconstrained trellis coded quantization for encoding the wavelet coeficients of both monochrome and color images. Excellent peak signal-to-noise ratios are obtained for encoding monochrome and color versionsof the 512 x 512 "Lenna" image. Comparisons with other results from the literature reveal that the proposed wavelet coder is quite competitive.

The feasibility study on carbon fibre reinforced polymer (CFRP) fabrics in axial strengthening of hollow square sections (HSS) was investigated in this paper. CFRP was used as strips form with other parameters such as the number of layers and spacing of strips. Experimental results revealed that the external bonding of normal modulus CFRP strips significantly enhanced the load carrying capacity and stiffness of the hollow sections and also reduced the axial shortening of columns by providing external confinement against the elastic deformation. The increase in the CFRP strips thickness effectively delayed the local buckling of the above members and led to the inward buckling rather than outward one. Finally, three-dimensional nonlinear finite element modeling of CFRP strengthened hollow square sectionswas created by using ANSYS 12.0 to validate the results and the numerical results such as failure modes and load deformation behaviour fairly agreed with the experimental results.

The discrete wavelet transform has recently emerged as a powerful technique for decomposing images into various multi-resolution approximations. Multi-resolution decomposition schemes have proven to be very effective for high-quality, low bit-rate image coding. In this work, we investigate the use of entropy-constrained trellis-coded quantization (ECTCQ) for encoding the wavelet coefficients of both monochrome and color images. ECTCQ is known as an effective scheme for quantizing memoryless sources with low to moderate complexity, The ECTCQ approach to data compression has led to some of the most effective source codes found to date for memoryless sources. Performance comparisons are made using the classical quadrature mirror filter bank of Johnston and nine-tap spline filters that were built from biorthogonal wavelet bases. We conclude that the encoded images obtained from the system employing nine-tap spline filters are marginally superior although at the expense of additional computational burden. Excellent peak-signal-to-noise ratios are obtained for encoding monochrome and color versions of the 512x512 "Lenna" image. Comparisons with other results from the literature reveal that the proposed wavelet coder is quite competitive.

The application of transform coding using trellis coded quantization (TCQ) to encode monochrome and color images is investigated. Specifically, TCQ is used to encode transform coefficients resulting from applying a 16 X 16 discrete cosine transform (DCT) to 8-b gray level and 24-b color images. For the color images, the red, green, and blue (RGB) planes are transformed into the NTSC transmission primaries (Y, I, and Q) before the DCT is applied. Rate allocation schemes for encoding the transform coefficients and side information are presented and discussed. The generalized Lloyd algorithm and the entropy-constrained vector quantizer design algorithm are used to design fixed-rate and entropy-constrained codebooks. The entropy-constrained designs result in a low-complexity system with objective performance rivaling that of many coding schemes from the literature.