Search

Object re-identification (ReID) in large camera networks faces numerous challenges. First, the similar appearances of objects degrade ReID performance, a challenge that needs to be addressed by existing appearance-based ReID methods. Second, most ReID studies are performed in laboratory settings and do not consider real-world scenarios. To overcome these challenges, we introduce a novel ReID framework that leverages a spatial-temporal fusion network and causal identity matching (CIM). Our framework estimates camera network topology using a proposed adaptive Parzen window and combines appearance features with spatial-temporal cues within the fusion network. This approach has demonstrated outstanding performance across several datasets, including VeRi776, Vehicle-3I, and Market-1501, achieving up to 99.70% rank-1 accuracy and 95.5% mAP. Furthermore, the proposed CIM approach, which dynamically assigns gallery sets based on camera network topology, has further improved ReID accuracy and robustness in real-world settings, evidenced by a 94.95% mAP and a 95.19% F1 score on the Vehicle-3I dataset. The experimental results support the effectiveness of incorporating spatial-temporal information and CIM for real-world ReID scenarios, regardless of the data domain (e.g., vehicle, person).

Session-based recommendation (SBR) aims to predict the following item a user will interact with during an ongoing session. Most existing SBR models focus on designing sophisticated neural-based encoders to learn a session representation, capturing the relationship among session items. However, they tend to focus on the last item, neglecting diverse user intents that may exist within a session. This limitation leads to significant performance drops, especially for longer sessions. To address this issue, we propose a novel SBR model, called Multi-intent-aware Session-based Recommendation Model (MiaSRec). It adopts frequency embedding vectors indicating the item frequency in session to enhance the information about repeated items. MiaSRec represents various user intents by deriving multiple session representations centered on each item and dynamically selecting the important ones. Extensive experimental results show that MiaSRec outperforms existing state-of-the-art SBR models on six datasets, particularly those with longer average session length, achieving up to 6.27% and 24.56% gains for MRR@20 and Recall@20. Our code is available at https://github.com/jin530/MiaSRec., Comment: SIGIR 2024. 5 pages

Text role classification involves classifying the semantic role of textual elements within scientific charts. For this task, we propose to finetune two pretrained multimodal document layout analysis models, LayoutLMv3 and UDOP, on chart datasets. The transformers utilize the three modalities of text, image, and layout as input. We further investigate whether data augmentation and balancing methods help the performance of the models. The models are evaluated on various chart datasets, and results show that LayoutLMv3 outperforms UDOP in all experiments. LayoutLMv3 achieves the highest F1-macro score of 82.87 on the ICPR22 test dataset, beating the best-performing model from the ICPR22 CHART-Infographics challenge. Moreover, the robustness of the models is tested on a synthetic noisy dataset ICPR22-N. Finally, the generalizability of the models is evaluated on three chart datasets, CHIME-R, DeGruyter, and EconBiz, for which we added labels for the text roles. Findings indicate that even in cases where there is limited training data, transformers can be used with the help of data augmentation and balancing methods. The source code and datasets are available on GitHub under https://github.com/hjkimk/text-role-classification

It is known from archival TESS data that the semi-detached Algol system XZ UMa is one of the candidate binary stars exhibiting short-period oscillations. We secured new high-resolution spectroscopic observations for the program target to better understand its binary and pulsation properties. From the echelle spectra, the radial velocities (RVs) of the eclipsing pair were derived, and the atmosphere parameters of the primary component were measured to be $v_{\rm A}\sin$$i = 80\pm7$ km s$^{-1}$, $T_{\rm eff,A}$ = $7940\pm120$ K, and [M/H] = $-0.15\pm0.20$. The combined solution of our double-lined RVs and the TESS data provides robust physical parameters for XZ UMa with mass and radius measurement precision of better than 2 \%. The outside-eclipse residuals from a mean light curve in the 0.002 phase bin were used for multifrequency analyses, and we extracted 32 significant frequencies (22 in $<$ 5.0 day$^{-1}$ and 10 in 39$-$52 day$^{-1}$). The low frequencies may be mostly aliasing sidelobes, while six of the high frequencies may be pulsation signals arising from the detached primary located inside the $\delta$ Sct domain. Their periods, pulsation constants, and pulsational-orbital-period ratios indicate that the mass-accretion primary star is a $\delta$ Sct pulsator and, hence, XZ UMa is an oscillating eclipsing Algol., Comment: 25 pages, including 7 figures and 5 tables, accepted for publication in PASJ

We present detailed analyses of updated eclipse timing diagrams for 32 contact binary merger candidates in the Galactic bulge. The photometric data was obtained from 2016 to 2021 using the Korea Microlensing Telescope Network (KMTNet) with the 1.6 m telescopes located at three southern sites (CTIO, SAAO, and SSO). The times of minimum lights were determined by applying the binary star model to full light curves created at half-year intervals from the observations. The orbital period variations of the binary systems were analyzed using the $O-C$ diagrams from our new timings with the others published in the literature (Hong et al.), which are based on the OGLE observations from 2001 to 2015. As results, the orbital periods and period decreasing rates of 32 binary systems were located to be in the ranges of 0.370$-$1.238 days and from $-3.0$ to $-13.1\times10^{-6}$ day yr$^{-1}$, respectively. Out of these stars, 24 systems show a combination effect of a parabola and a light travel time caused by a third body and their outer orbital periods are in the range of 9.1$-$26.5 yr, respectively. We propose that all of our merger candidates need additional monitoring observations to study a luminous red nova (LRN) progenitor., Comment: 15 pages, 5 figures

Vehicle re-identification (ReID) in a large-scale camera network is important in public safety, traffic control, and security. However, due to the appearance ambiguities of vehicle, the previous appearance-based ReID methods often fail to track vehicle across multiple cameras. To overcome the challenge, we propose a spatial-temporal vehicle ReID framework that estimates reliable camera network topology based on the adaptive Parzen window method and optimally combines the appearance and spatial-temporal similarities through the fusion network. Based on the proposed methods, we performed superior performance on the public dataset (VeRi776) by 99.64% of rank-1 accuracy. The experimental results support that utilizing spatial and temporal information for ReID can leverage the accuracy of appearance-based methods and effectively deal with appearance ambiguities., Comment: 10 pages, 6 figures

The aim of this paper is to study the $\lambda$-Stirling numbers of both kinds which are $\lambda$-analogues of Stirling numbers of both kinds. Those numbers have nice combinatorial interpretations when $\lambda$ are positive integers. If $\lambda$ =1, then the $\lambda$-Stirling numbers of both kinds reduce to the Stirling numbers of both kinds. We derive new types of generating functions of the $\lambda$-Stirling numbers of both kinds which are related to the reciprocals of the generalized rising factorials. Furthermore, some related identities are also derived from those generating functions. In addition, all the corresponding results to the $\lambda$-Stirling numbers of both kinds are obtained also for the $\lambda$-analogues of r-Stirling numbers of both kinds which are generalizations of those numbers., Comment: 10 pages