Research Outputs

Permanent URI for this communityhttps://hdl.handle.net/20.500.14288/2

Browse

Filters

Advanced Search

Filter by

Settings

Sponsors: search.filters.sponsors.TÜBİTAKSubject: search.filters.subject.Computer engineering

Search Results

Now showing 1 - 10 of 25
  • Placeholder
    PublicationMetadata only
    3D articulated shape segmentation using motion information
    (Institute of Electrical and Electronics Engineers (IEEE), 2010) Department of Computer Engineering; N/A; Yemez, Yücel; Kalafatlar, Emre; Faculty Member; Master Student; Department of Computer Engineering; College of Engineering; Graduate School of Sciences and Engineering; 107907; N/A
    We present a method for segmentation of articulated 3D shapes by incorporating the motion information obtained from time-varying models. We assume that the articulated shape is given in the form of a mesh sequence with fixed connectivity so that the inter-frame vertex correspondences, hence the vertex movements, are known a priori. We use different postures of an articulated shape in multiple frames to constitute an affinity matrix which encodes both temporal and spatial similarities between surface points. The shape is then decomposed into segments in spectral domain based on the affinity matrix using a standard K-means clustering algorithm. The performance of the proposed segmentation method is demonstrated on the mesh sequence of a human actor.
  • Placeholder
    PublicationMetadata only
    Affect burst detection using multi-modal cues
    (IEEE, 2015) Department of Computer Engineering; Department of Computer Engineering; N/A; Department of Computer Engineering; N/A; Sezgin, Tevfik Metin; Yemez, Yücel; Türker, Bekir Berker; Erzin, Engin; Marzban, Shabbir; Faculty Member; Faculty Member; PhD Student; Faculty Member; Master Student; Department of Computer Engineering; College of Engineering; College of Engineering; Graduate School of Sciences and Engineering; College of Engineering; Graduate School of Sciences and Engineering; 18632; 107907; N/A; 34503; N/A
    Recently, affect bursts have gained significant importance in the field of emotion recognition since they can serve as prior in recognising underlying affect bursts. In this paper we propose a data driven approach for detecting affect bursts using multimodal streams of input such as audio and facial landmark points. The proposed Gaussian Mixture Model based method learns each modality independently followed by combining the probabilistic outputs to form a decision. This gives us an edge over feature fusion based methods as it allows us to handle events when one of the modalities is too noisy or not available. We demonstrate robustness of the proposed approach on 'Interactive emotional dyadic motion capture database' (IEMOCAP) which contains realistic and natural dyadic conversations. This database is annotated by three annotators to segment and label affect bursts to be used for training and testing purposes. We also present performance comparison between SVM based methods and GMM based methods for the same configuration of experiments.
  • Placeholder
    PublicationMetadata only
    Affect-expressive hand gestures synthesis and animation
    (IEEE, 2015) Department of Computer Engineering; N/A; Department of Computer Engineering; Erzin, Engin; Bozkurt, Elif; Yemez, Yücel; Faculty Member; PhD Student; Faculty Member; Department of Computer Engineering; College of Engineering; Graduate School of Sciences and Engineering; College of Engineering; 34503; N/A; 107907
    Speech and hand gestures form a composite communicative signal that boosts the naturalness and affectiveness of the communication. We present a multimodal framework for joint analysis of continuous affect, speech prosody and hand gestures towards automatic synthesis of realistic hand gestures from spontaneous speech using the hidden semi-Markov models (HSMMs). To the best of our knowledge, this is the first attempt for synthesizing hand gestures using continuous dimensional affect space, i.e., activation, valence, and dominance. We model relationships between acoustic features describing speech prosody and hand gestures with and without using the continuous affect information in speaker independent configurations and evaluate the multimodal analysis framework by generating hand gesture animations, also via objective evaluations. Our experimental studies are promising, conveying the role of affect for modeling the dynamics of speech-gesture relationship. © 2015 IEEE.
  • Placeholder
    PublicationMetadata only
    Analytical model for topology dependence in peer-to-peer anti-entropy spreading
    (Bogazici University, 2008) N/A; Department of Computer Engineering; Department of Mathematics; Özkasap, Öznur; Çağlar, Mine; İskender, Emre; Faculty Member; Faculty Member; Master Student; Department of Computer Engineering; Department of Mathematics; College of Engineering; College of Sciences; Graduate School of Sciences and Engineering; 113507; 105131; N/A
    We examine spreading of epidemics for an anti-entropy algorithm in networks with various P2P (peer-to-peer) overlay topologies. Neighborhood knowledge among peers and information exchange based on proximity are considered. Our analytical model for SI (Susceptible-Infected) epidemics involves equations for calculating the infection probability of each peer in consecutive epidemic rounds as a function of the topology. Using numerical evaluations, we study the effect of graph properties on dissemination as an aspect of real world P2P overlays
  • Thumbnail Image
    PublicationOpen Access
    Characterizing user behavior for speech and sketch-based video retrieval interfaces
    (Association for Computing Machinery (ACM), 2017) Department of Computer Engineering; Sezgin, Tevfik Metin; Altıok, Ozan Can; Faculty Member; Master Student; Department of Computer Engineering; College of Engineering; 18632; N/A
    From a user interaction perspective, speech and sketching make a good couple for describing motion. Speech allows easy specification of content, events and relationships, while sketching brings in spatial expressiveness. Yet, we have insufficient knowledge of how sketching and speech can be used for motion-based video retrieval, because there are no existing retrieval systems that support such interaction. In this paper, we describe a Wizard-of-Oz protocol and a set of tools that we have developed to engage users in a sketch-and speech-based video retrieval task. We report how the tools and the protocol fit together using "retrieval of soccer videos" as a use case scenario. Our software is highly customizable, and our protocol is easy to follow. We believe that together they will serve as a convenient and powerful duo for studying a wide range of multi-modal use cases.
  • Placeholder
    PublicationMetadata only
    Comparison of phoneme and viseme based acoustic units for speech driven realistic lip animation
    (IEEE, 2007) Bozkurt, Elif; Erdem, Çiǧdem Eroǧlu; Erdem, Tanju; Özkan, Mehmet; Department of Computer Engineering; Erzin, Engin; Faculty Member; Department of Computer Engineering; College of Engineering; 34503
    Natural looking lip animation, synchronized with incoming speech, is essential for realistic character animation. In this work, we evaluate the performance of phone and viseme based acoustic units, with and without context information, for generating realistic lip synchronization using HMM based recognition systems. We conclude via objective evaluations that utilization of viseme based units with context information outperforms the other methods./ Öz: Konuşma ile senkronize ve doğal görünen dudak hareketlerinin üretilmesi, gerçekçi karakter animasyonu için önemli bir problemdir. Bu çalışmada, gerçekçi dudak hareketleri üretebilmek için Saklı Markov Modeli (SMM) kullanarak, fonem ve vizem temelli akustik birimlerin başarımlarını karşılaştırıyoruz. Nesnel değerlendirmeler sonucunda, komşuluk bilgisini kullanan vizem temelli akustik birimlerin diğer metodlardan daha üstün olduğunu gösteriyoruz.
  • Placeholder
    PublicationMetadata only
    Distributed key selection for group applications in ad Hoc networks
    (IEEE, 2008) N/A; Department of Computer Engineering; Özkasap, Öznur; Obut, Esra; Faculty Member; Undergraduated Student; Department of Computer Engineering; College of Engineering; Graduate School of Sciences and Engineering; 113507; N/A
    Efficient key management is an indispensable element for supporting security in ad hoc networks. However, several key pre-distribution approaches based on a trusted third party have limitations caused by their centralized principles. Distributed key selection mechanisms are offered to deal with the limitations of the centralized counterparts in ad hoc networks. In this study, we consider distributed key selection for ad hoc networked group applications. We provide extensions to a recent key establishment approach named SeeDKS (Seed-based Distributed Key Selection). The approach is generalized by offering a novel exclusion property testing required for secure communications. Secure group management algorithms for SeeDKS are also proposed. Quantitative evaluation and superiority of the generalized algorithm over distributed key selection are presented with experimental results.
  • Placeholder
    PublicationMetadata only
    Energy efficient hierarchical epidemics in peer-to-peer systems
    (IEEE, 2011) N/A; Department of Computer Engineering; Özkasap, Öznur; Çem, Emrah; Koç, Tuğba; Faculty Member; PhD Student; Master Student; Department of Computer Engineering; College of Engineering; Graduate School of Sciences and Engineering; Graduate School of Sciences and Engineering; 113507; N/A; N/A
    Epidemic or gossip-based mechanisms are preferred in several distributed protocols for their ease of deployment, simplicity, robustness against failures, load-balancing and limited resource usage. In flat neighborhood epidemics, peers have similar responsibilities and all participate in gossiping via neighboring peers. We have proposed an energy cost model for a generic peer using flat neighborhood epidemics, and examined the effect of protocol parameters to characterize energy consumption. Although it has been shown that a peers power consumption amount is independent of population size, peers always need to be active to process incoming gossip messages. In this study, we consider power awareness features of flat and hierarchical epidemics in peer-to-peer (P2P) systems, and propose a power-aware hierarchical epidemic approach with its energy cost model and analysis. In this adaptive approach, only a subset of peer population is active in gossiping by forming an overlay, so that the other peers can switch to idle state. It also allows data aggregation that can be utilized to reduce gossip message size. As a case study for epidemic protocol, we use our approach and simulation model for frequent item set discovery in unstructured P2P networks.
  • Placeholder
    PublicationMetadata only
    Energy efficient video decoding on multi-core devices
    (ACM, 2012) N/A; Department of Computer Engineering; Department of Electrical and Electronics Engineering; N/A; N/A; Özkasap, Öznur; Tekalp, Ahmet Murat; Gürler, Cihat Göktuğ; Kılıçarslan, Damla; Faculty Member; Faculty Member; PhD Student; Master Student; Department of Computer Engineering; Department of Electrical and Electronics Engineering; College of Engineering; College of Engineering; Graduate School of Sciences and Engineering; Graduate School of Sciences and Engineering; 113507; 26207; N/A; N/A
    Emergence of high quality media applications results in larger data sizes as well as higher bitrates of digital multimedia contents, and their significant share on the overall Internet traffic. These lead to an increase in the energy consumption rates and performance requirements for real-time video decoding. In this study, we propose parallel video decoding solutions to provide real-time decoding performance with reduced energy consumption on multi-core devices. Various approaches of parallelism at data and task levels can be incorporated in video decoders, bringing efficiency in energy consumption rates and/or performance. We offer and develop two approaches for the H.264 standard. The former is based on a coarse-grained frame level, and the latter is a fine-grained macroblock level parallelism. The implementations were conducted on a shared memory multi-core platform as an all software solution for real-time scalable video decoding. We also discuss energy efficiency as well as performance results. As part of our ongoing work, further parallelization methods such as parallelism at slice level, and parallel decoding of consecutive groups of pictures on the H.264/SVC decoder are discussed
  • Placeholder
    PublicationMetadata only
    Estimation of acoustic microphone vocal tract parameters from throat microphone recordings
    (IEEE, 2007) Department of Computer Engineering; N/A; Erzin, Engin; Akargün, Ülkü Çağrı; Faculty Member; Master Student; Department of Computer Engineering; College of Engineering; Graduate School of Sciences and Engineering; 34503; N/A
    Recently, joint processing of throat and acoustic microphone recordings has been an attractive tool for robust speech processing. As the throat microphones record the acoustic sounds in the form of vibrations from skin attached sensors, they are more robust and highly correlated with the acoustic speech signal. We investigate the correlation of throat and acoustic microphone recordings. We propose a hidden Markov model (HMM) based structure to estimate acoustic speech features from throat speech features. The HMM based estimator will be used to estimate clean acoustic speech features from noisy throat and acoustic microphone recordings. Experimental results on acoustic speech feature estimation are provided./ Öz: Gırtlak mikrofonları akustik mikrofon ile yüksek ilintili ve ortam gürültüsünden az etkilenir olmaları nedeni ile gürbüz konuşma işleme uygulamalarında kullanılmaktadır. Bu çalışmada gırtlak mikrofonlarının akustik ses mikrofonları ile olan ilintisini araştırdık. Gırtlak mikrofonu sesi doku üzerindeki titreşimlerden kaydettiği için değişken çevre şartlarına daha gürbüz ama asıl ses işaretinden daha düşük ve bozunmuş bir bant genliğine sahiptir. Gırtlak mikrofonu ve akustik mikrofon ilintisini kullanarak gırtlak mikrofonu ses özniteliklerinden akustik mikrofon ses özniteliklerini kestirmek için saklı Markov modellerine dayalı bir yapı önerdik. Bu kestiriciyi gürültü altında gırtlak ve gürültülü akustik modellerden, temiz akustik modellerin kestirimi için kullanacağız. Akustik mikrofon ses özniteliklerinin kestirimi üzerine sonuçlar sunulmuştur.