Showing 1–12 of 12 results for author: T, G

SAM-PM: Enhancing Video Camouflaged Object Detection using Spatio-Temporal Attention

Authors: Muhammad Nawfal Meeran, Gokul Adethya T, Bhanu Pratyush Mantha

Abstract: In the domain of large foundation models, the Segment Anything Model (SAM) has gained notable recognition for its exceptional performance in image segmentation. However, tackling the video camouflage object detection (VCOD) task presents a unique challenge. Camouflaged objects typically blend into the background, making them difficult to distinguish in still images. Additionally, ensuring temporal… ▽ More In the domain of large foundation models, the Segment Anything Model (SAM) has gained notable recognition for its exceptional performance in image segmentation. However, tackling the video camouflage object detection (VCOD) task presents a unique challenge. Camouflaged objects typically blend into the background, making them difficult to distinguish in still images. Additionally, ensuring temporal consistency in this context is a challenging problem. As a result, SAM encounters limitations and falls short when applied to the VCOD task. To overcome these challenges, we propose a new method called the SAM Propagation Module (SAM-PM). Our propagation module enforces temporal consistency within SAM by employing spatio-temporal cross-attention mechanisms. Moreover, we exclusively train the propagation module while kee** the SAM network weights frozen, allowing us to integrate task-specific insights with the vast knowledge accumulated by the large model. Our method effectively incorporates temporal consistency and domain-specific expertise into the segmentation network with an addition of less than 1% of SAM's parameters. Extensive experimentation reveals a substantial performance improvement in the VCOD benchmark when compared to the most recent state-of-the-art techniques. Code and pre-trained weights are open-sourced at https://github.com/SpiderNitt/SAM-PM △ Less

Submitted 9 June, 2024; originally announced June 2024.

Beyond Uniform Scaling: Exploring Depth Heterogeneity in Neural Architectures

Authors: Akash Guna R. T, Arnav Chavan, Deepak Gupta

Abstract: Conventional scaling of neural networks typically involves designing a base network and growing different dimensions like width, depth, etc. of the same by some predefined scaling factors. We introduce an automated scaling approach leveraging second-order loss landscape information. Our method is flexible towards skip connections a mainstay in modern vision transformers. Our training-aware method… ▽ More Conventional scaling of neural networks typically involves designing a base network and growing different dimensions like width, depth, etc. of the same by some predefined scaling factors. We introduce an automated scaling approach leveraging second-order loss landscape information. Our method is flexible towards skip connections a mainstay in modern vision transformers. Our training-aware method jointly scales and trains transformers without additional training iterations. Motivated by the hypothesis that not all neurons need uniform depth complexity, our approach embraces depth heterogeneity. Extensive evaluations on DeiT-S with ImageNet100 show a 2.5% accuracy gain and 10% parameter efficiency improvement over conventional scaling. Scaled networks demonstrate superior performance upon training small scale datasets from scratch. We introduce the first intact scaling mechanism for vision transformers, a step towards efficient model scaling. △ Less

Submitted 19 February, 2024; originally announced February 2024.

Comments: Accepted At ICLR 2024 (Tiny Paper Track)

SN 2022jli: a type Ic supernova with periodic modulation of its light curve and an unusually long rise

Authors: Moore T., Smartt S. J., Nicholl M., Srivastav S., Stevance H. F., Jess D. B., Grant S. D. T., Fulton M. D., Rhodes L., Sim S. A., Hirai R., Podsiadlowski P., Anderson J. P., Ashall C., Bate W., Fender R., Gutierrez C. P., Howell D. A., Huber M. E., Inserra C., Leloudas G., Monard L. A. G., Muller-Bravo T. E., Shappee B. J., Smith K. W. , et al. (20 additional authors not shown)

Abstract: We present multi-wavelength photometry and spectroscopy of SN 2022jli, an unprecedented Type Ic supernova discovered in the galaxy NGC 157 at a distance of $\approx$ 23 Mpc. The multi-band light curves reveal many remarkable characteristics. Peaking at a magnitude of $g=15.11\pm0.02$, the high-cadence photometry reveals 12.5$\pm0.2\ $day periodic undulations superimposed on the 200 day supernova d… ▽ More We present multi-wavelength photometry and spectroscopy of SN 2022jli, an unprecedented Type Ic supernova discovered in the galaxy NGC 157 at a distance of $\approx$ 23 Mpc. The multi-band light curves reveal many remarkable characteristics. Peaking at a magnitude of $g=15.11\pm0.02$, the high-cadence photometry reveals 12.5$\pm0.2\ $day periodic undulations superimposed on the 200 day supernova decline. This periodicity is observed in the light curves from nine separate filter and instrument configurations with peak-to-peak amplitudes of $\simeq$ 0.1 mag. This is the first time that repeated periodic oscillations, over many cycles, have been detected in a supernova light curve. SN 2022jli also displays an extreme early excess which fades over $\approx$ 25 days followed by a rise to a peak luminosity of $L_{\rm opt} = 10^{42.1}$ erg s$^{-1}$. Although the exact explosion epoch is not constrained by data, the time from explosion to maximum light is $\gtrsim$ 59 days. The luminosity can be explained by a large ejecta mass ($M_{\rm ej}\approx12\pm6$M$_{\odot}$) powered by $^{56}$Ni but we find difficulty in quantitatively modelling the early excess with circumstellar interaction and cooling. Collision between the supernova ejecta and a binary companion is a possible source of this emission. We discuss the origin of the periodic variability in the light curve, including interaction of the SN ejecta with nested shells of circumstellar matter and neutron stars colliding with binary companions. △ Less

Submitted 22 September, 2023; originally announced September 2023.

Comments: Accepted in ApJL

Do** of large amount tetravalent Ge ions into Fe2O3 structure and experimental results on modified structural, optical and electronic properties

Authors: Divya Sherin G T, R. N Bhowmik

Abstract: We report the do** high concentration of tetravalent Ge4+ ions (5 mol % for x = 0.05 to 30 mol % for x = 0.30) at the Fe3+ sites of Fe2-xGexO3 system by chemical coprecipitation route. The charge state of Fe and Ge ions has been modified into lower values, in addition to their normal +3 and +4 states, to stabilize the rhombohedral phase of hematite (α-Fe2O3) structure. X-ray photoelectron spectr… ▽ More We report the do** high concentration of tetravalent Ge4+ ions (5 mol % for x = 0.05 to 30 mol % for x = 0.30) at the Fe3+ sites of Fe2-xGexO3 system by chemical coprecipitation route. The charge state of Fe and Ge ions has been modified into lower values, in addition to their normal +3 and +4 states, to stabilize the rhombohedral phase of hematite (α-Fe2O3) structure. X-ray photoelectron spectra and optical band gap measurements indicated a combination of ionic and covalence character of metal-oxygen bonds as an effect of Ge do** in hematite structure. The Ge doped hematite samples have exhibited wide band gap semiconductor properties with band gap 4.50-4.70 eV and remarkably enhanced electrical conductivity (σ ~ 10-4 S/m) in comparison to α-Fe2O3 (10-11 S/m). The thermo-conductivity measurements using warming and cooling modes showed a highly irreversible feature in the semiconductor regime at higher temperatures. Some of the samples indicated metal-like state at lower temperature, in addition to semiconductor state. Our experimental results confirmed the strategy of enhancing electrical conductivity by do** tetravalent ions in hematite structure. It has been understood that combination of ionic and covalence character of the metal-oxygen bonds has played an important role in modifying the semiconductor properties in Ge doped Fe2O3 system. △ Less

Submitted 7 September, 2023; originally announced September 2023.

Comments: 34 pages, 7 Figures, Tables 4

The MUSE second-generation VLT instrument

Authors: Bacon R., Accardo M., Adjali L., Anwand H., Bauer S., Biswas I., Blaizot J., Boudon D., Brau-Nogue S., Brinchmann J., Caillier P., Capoani L., Carollo C. M., Contini T., Couderc P., Daguise E., Deiries S., Delabre B., Dreizler S., Dubois J. P., Dupieux M., Dupuy C., Emsellem E., Fechner T., Fleischmann A. , et al. (43 additional authors not shown)

Abstract: The Multi Unit Spectroscopic Explorer (MUSE) is a second-generation VLT panoramic integral-field spectrograph currently in manufacturing, assembly and integration phase. MUSE has a field of 1x1 arcmin2 sampled at 0.2x0.2 arcsec2 and is assisted by the VLT ground layer adaptive optics ESO facility using four laser guide stars. The instrument is a large assembly of 24 identical high performance inte… ▽ More The Multi Unit Spectroscopic Explorer (MUSE) is a second-generation VLT panoramic integral-field spectrograph currently in manufacturing, assembly and integration phase. MUSE has a field of 1x1 arcmin2 sampled at 0.2x0.2 arcsec2 and is assisted by the VLT ground layer adaptive optics ESO facility using four laser guide stars. The instrument is a large assembly of 24 identical high performance integral field units, each one composed of an advanced image slicer, a spectrograph and a 4kx4k detector. In this paper we review the progress of the manufacturing and report the performance achieved with the first integral field unit. △ Less

Submitted 30 November, 2022; originally announced November 2022.

Comments: 10 pages, 5 figures

Fast-Image2Point: Towards Real-Time Point Cloud Reconstruction of a Single Image using 3D Supervision

Authors: AmirHossein Zamani, Amir G. Aghdam, Kamran Ghaffari T

Abstract: A key question in the problem of 3D reconstruction is how to train a machine or a robot to model 3D objects. Many tasks like navigation in real-time systems such as autonomous vehicles directly depend on this problem. These systems usually have limited computational power. Despite considerable progress in 3D reconstruction systems in recent years, applying them to real-time systems such as navigat… ▽ More A key question in the problem of 3D reconstruction is how to train a machine or a robot to model 3D objects. Many tasks like navigation in real-time systems such as autonomous vehicles directly depend on this problem. These systems usually have limited computational power. Despite considerable progress in 3D reconstruction systems in recent years, applying them to real-time systems such as navigation systems in autonomous vehicles is still challenging due to the high complexity and computational demand of the existing methods. This study addresses current problems in reconstructing objects displayed in a single-view image in a faster (real-time) fashion. To this end, a simple yet powerful deep neural framework is developed. The proposed framework consists of two components: the feature extractor module and the 3D generator module. We use point cloud representation for the output of our reconstruction module. The ShapeNet dataset is utilized to compare the method with the existing results in terms of computation time and accuracy. Simulations demonstrate the superior performance of the proposed method. Index Terms-Real-time 3D reconstruction, single-view reconstruction, supervised learning, deep neural network △ Less

Submitted 20 September, 2022; originally announced September 2022.

Comments: 1- The paper has 8 pages 2- The paper has already been accepted in IEEE International Conference on Machine Learning and Applications (ICMLA) 2022 and it will appear in IEEE ICMLA 2022 proceedings

Three new brown dwarfs and a massive hot Jupiter revealed by TESS around early-type stars

Authors: Psaridi A., Bouchy F., Lendl M., Grieves N., Stassun K. G., Carmichael T., Gill S., Peña Rojas P. A., Gan T., Shporer A., Bieryla A., Christiansen J. L, Crossfield I. J. M, Galland F. Hooton M. J. Jenkins J. M, Jenkins J. S, Latham D. W, Lund M. B, Rodriguez J. E, Ting E. B, Udry S. Ulmer-Moll S. Wittenmyer R. A, Yanzhe Zhang Y., Zhou G., Addison B., Cointepas M., Collins K. A. , et al. (18 additional authors not shown)

Abstract: The detection and characterization of exoplanets and brown dwarfs (BDs) around massive AF-type stars is essential to investigate and constrain the impact of stellar mass on planet properties. However, such targets are still poorly explored in radial velocity (RV) surveys because they only feature a small number of stellar lines and those are usually broadened and blended by stellar rotation as wel… ▽ More The detection and characterization of exoplanets and brown dwarfs (BDs) around massive AF-type stars is essential to investigate and constrain the impact of stellar mass on planet properties. However, such targets are still poorly explored in radial velocity (RV) surveys because they only feature a small number of stellar lines and those are usually broadened and blended by stellar rotation as well as stellar jitter. As a result, the available information about the formation and evolution of planets and BDs around hot stars is limited. We aim to increase the sample and precisely measure the masses and eccentricities of giant planets and BDs transiting AF-type stars detected by the Transiting Exoplanet Survey Satellite (TESS). We followed bright (V < 12 mag) stars with $T_{\mathrm{eff}}$ > 6200 K that host giant companions (R > 7 $\mathrm{R_{\rm \oplus}}$) using ground-based photometric observations as well as high precision RV measurements from the CORALIE, CHIRON, TRES, FEROS, and MINERVA-Australis spectrographs. In the context, we present the discovery of three BD companions, TOI-629b, TOI-1982b, and TOI-2543b, and one massive planet, TOI-1107b. From the joint analysis we find the BDs have masses between 66 and 68 $\mathrm{M_{\rm Jup}}$, periods between 7.54 and 17.17 days, and radii between 0.95 and 1.11 $\mathrm{R_{\rm Jup}}$. The hot Jupiter TOI-1107b has an orbital period of 4.08 days, a radius of 1.30 $\mathrm{R_{\rm Jup}}$, and a mass of 3.35 $\mathrm{M_{\rm Jup}}$. As a by-product of this program, we identified four low-mass eclipsing components (TOI-288b, TOI-446b, TOI-478b, and TOI-764b). Both TOI-1107b and TOI-1982b present an anomalously inflated radius with respect to the age of these systems. TOI-629 is among the hottest stars with a known transiting brown dwarf. TOI-629b and TOI-1982b are among the most eccentric brown dwarfs. △ Less

Submitted 22 May, 2022; originally announced May 2022.

Comments: 21 pages, 13 figures, Accepted by Astronomy & Astrophysics

Journal ref: A&A 664, A94 (2022)

Local heating due to convective overshooting and the solar modelling problem

Authors: Baraffe I, Constantino T, Clarke J, Le Saux A, Goffrey T, Guillet T, Pratt J, Vlaykov D. G

Abstract: Recent hydrodynamical simulations of convection in a solar-like model suggest that penetrative convective flows at the boundary of the convective envelope modify the thermal background in the overshooting layer. Based on these results, we implement in one-dimensional stellar evolution codes a simple prescription to modify the temperature gradient below the convective boundary of a solar model. Thi… ▽ More Recent hydrodynamical simulations of convection in a solar-like model suggest that penetrative convective flows at the boundary of the convective envelope modify the thermal background in the overshooting layer. Based on these results, we implement in one-dimensional stellar evolution codes a simple prescription to modify the temperature gradient below the convective boundary of a solar model. This simple prescription qualitatively reproduces the behaviour found in the hydrodynamical simulations, namely a local heating and smoothing of the temperature gradient below the convective boundary. We show that introducing local heating in the overshooting layer can reduce the sound-speed discrepancy usually reported between solar models and the structure of the Sun inferred from helioseismology. It also affects key quantities in the convective envelope, such as the density, the entropy, and the speed of sound. These effects could help reduce the discrepancies between solar models and observed constraints based on seismic inversions of the Ledoux discriminant. Since mixing due to overshooting and local heating are the result of the same convective penetration process, the goal of this work is to invite solar modellers to consider both processes for a more consistent approach. △ Less

Submitted 1 January, 2022; originally announced January 2022.

Comments: 7 pages, 4 figures, accepted for publication in A&A

Journal ref: A&A 659, A53 (2022)

Space Photometry with BRITE-Constellation

Authors: Weiss W. W, Zwintz K., Kuschnig R., Handler G., Moffat A. F. J., Baade D., Bowman D. M., Granzer T., Kallinger T., Koudelka O. F., Lovekin C. C., Neiner C., Pablo H., Pigulski A., Popowicz A., Ramiaramanantsoa T., Rucinski S. M., Strassmeier K. G., Wade G. A

Abstract: BRITE-Constellation is devoted to high-precision optical photometric monitoring of bright stars, distributed all over the Milky Way, in red and/or blue passbands. Photometry from space avoids the turbulent and absorbing terrestrial atmosphere and allows for very long and continuous observing runs with high time resolution and thus provides the data necessary for understanding various processes ins… ▽ More BRITE-Constellation is devoted to high-precision optical photometric monitoring of bright stars, distributed all over the Milky Way, in red and/or blue passbands. Photometry from space avoids the turbulent and absorbing terrestrial atmosphere and allows for very long and continuous observing runs with high time resolution and thus provides the data necessary for understanding various processes inside stars (e.g., asteroseismology) and in their immediate environment. While the first astronomical observations from space focused on the spectral regions not accessible from the ground it soon became obvious around 1970 that avoiding the turbulent terrestrial atmosphere significantly improved the accuracy of photometry and satellites explicitly dedicated to high-quality photometry were launched. A perfect example is BRITE-Constellation, which is the result of a very successful cooperation between Austria, Canada and Poland. Research highlights for targets distributed nearly over the entire HRD are presented, but focus primarily on massive and hot stars. △ Less

Submitted 24 June, 2021; originally announced June 2021.

Comments: BRITE-Constellation was designed, built, launched, and is operated and supported by the Austrian Research Promotion Agency, the University of Vienna, the Technical University of Graz, the University of Innsbruck, the Canadian Space Agency, the University of Toronto Institute for Aerospace Studies, the Foundation for Polish Science and Technology and the National Science Centre. 23 figures

Journal ref: Universe 2021, 7, 199

NAOMI: the adaptive optics system of the Auxiliary Telescopes of the VLTI

Authors: Woillez J., Abad J. A., Abuter R., Aller Carpentier E., Alonso J., Andolfato L., Barriga P., Berger J. -P., Beuzit J. -L., Bonnet H., Bourdarot G., Bourget P., Brast R., Caniguante L., Cottalorda E., Darré P., Delabre B., Delboulbé A., Delplancke-Ströbele F., Dembet R., Donaldson R., Dorn R., Dupeyron J., Dupuy C., Egner S. , et al. (58 additional authors not shown)

Abstract: The tip-tilt stabilisation system of the 1.8 m Auxiliary Telescopes of the Very Large Telescope Interferometer was never dimensioned for robust fringe tracking, except when atmospheric seeing conditions are excellent. Increasing the level of wavefront correction at the telescopes is expected to improve the coupling into the single-mode fibres of the instruments, and enable robust fringe tracking e… ▽ More The tip-tilt stabilisation system of the 1.8 m Auxiliary Telescopes of the Very Large Telescope Interferometer was never dimensioned for robust fringe tracking, except when atmospheric seeing conditions are excellent. Increasing the level of wavefront correction at the telescopes is expected to improve the coupling into the single-mode fibres of the instruments, and enable robust fringe tracking even in degraded conditions. We deployed a new adaptive optics module for interferometry (NAOMI) on the Auxiliary Telescopes. We present its design, performance, and effect on the observations that are carried out with the interferometric instruments. △ Less

Submitted 19 August, 2019; originally announced August 2019.

Comments: 10 pages, 18 figures, 2 tables, A&A forthcoming

Context Specific Event Model For News Articles

Authors: Kowcika A, Uma Maheswari, Geetha T V

Abstract: We present a new context based event indexing and event ranking model for News Articles. The context event clusters formed from the UNL Graphs uses the modified scoring scheme for segmenting events which is followed by clustering of events. From the context clusters obtained three models are developed- Identification of Main and Sub events; Event Indexing and Event Ranking. Based on the properties… ▽ More We present a new context based event indexing and event ranking model for News Articles. The context event clusters formed from the UNL Graphs uses the modified scoring scheme for segmenting events which is followed by clustering of events. From the context clusters obtained three models are developed- Identification of Main and Sub events; Event Indexing and Event Ranking. Based on the properties considered from the UNL Graphs for the modified scoring main events and sub events associated with main-events are identified. The temporal details obtained from the context cluster are stored using hashmap data structure. The temporal details are place-where the event took; person-who involved in that event; time-when the event took place. Based on the information collected from the context clusters three indices are generated- Time index, Person index, and Place index. This index gives complete details about every event obtained from context clusters. A new scoring scheme is introduced for ranking the events. The scoring scheme for event ranking gives weight-age based on the priority level of the events. The priority level includes the occurrence of the event in the title of the document, event frequency, and inverse document frequency of the events. △ Less

Submitted 5 August, 2013; originally announced August 2013.

Comparison Of Modified Dual Ternary Indexing And Multi-Key Hashing Algorithms For Music Information Retrieval

Authors: Rajeswari Sridhar, A. Amudha, S. Karthiga, Geetha T V

Abstract: In this work we have compared two indexing algorithms that have been used to index and retrieve Carnatic music songs. We have compared a modified algorithm of the Dual ternary indexing algorithm for music indexing and retrieval with the multi-key hashing indexing algorithm proposed by us. The modification in the dual ternary algorithm was essential to handle variable length query phrase and to acc… ▽ More In this work we have compared two indexing algorithms that have been used to index and retrieve Carnatic music songs. We have compared a modified algorithm of the Dual ternary indexing algorithm for music indexing and retrieval with the multi-key hashing indexing algorithm proposed by us. The modification in the dual ternary algorithm was essential to handle variable length query phrase and to accommodate features specific to Carnatic music. The dual ternary indexing algorithm is adapted for Carnatic music by segmenting using the segmentation technique for Carnatic music. The dual ternary algorithm is compared with the multi-key hashing algorithm designed by us for indexing and retrieval in which features like MFCC, spectral flux, melody string and spectral centroid are used as features for indexing data into a hash table. The way in which collision resolution was handled by this hash table is different than the normal hash table approaches. It was observed that multi-key hashing based retrieval had a lesser time complexity than dual-ternary based indexing The algorithms were also compared for their precision and recall in which multi-key hashing had a better recall than modified dual ternary indexing for the sample data considered. △ Less

Submitted 29 July, 2010; originally announced July 2010.

Comments: 11 pages, 5 figures

Journal ref: International Journal of Artificial Intelligence & Applications 1.3 (2010) 59-69