Zobrazeno 1 - 10
of 366
pro vyhledávání: '"Govindaraju, Venu"'
Reconstructing 3D faces with facial geometry from single images has allowed for major advances in animation, generative models, and virtual reality. However, this ability to represent faces with their 3D features is not as fully explored by the facia
Externí odkaz:
http://arxiv.org/abs/2408.16907
A "match cut" is a common video editing technique where a pair of shots that have a similar composition transition fluidly from one to another. Although match cuts are often visual, certain match cuts involve the fluid transition of audio, where soun
Externí odkaz:
http://arxiv.org/abs/2408.10998
Autor:
Fedorishin, Dennis, Forte III, Livio, Schneider, Philip, Setlur, Srirangaraj, Govindaraju, Venu
Sound event detection (SED) is an active area of audio research that aims to detect the temporal occurrence of sounds. In this paper, we apply SED to engine fault detection by introducing a multimodal SED framework that detects fine-grained engine fa
Externí odkaz:
http://arxiv.org/abs/2403.11037
We present a comprehensive study of chart visual question-answering(QA) task, to address the challenges faced in comprehending and extracting data from chart visualizations within documents. Despite efforts to tackle this problem using synthetic char
Externí odkaz:
http://arxiv.org/abs/2308.01979
We introduce a novel bottom-up approach for the extraction of chart data. Our model utilizes images of charts as inputs and learns to detect keypoints (KP), which are used to reconstruct the components within the plot area. Our novelty lies in detect
Externí odkaz:
http://arxiv.org/abs/2308.01971
Autor:
Jawade, Bhavin, Mohan, Deen Dayal, Fedorishin, Dennis, Setlur, Srirangaraj, Govindaraju, Venu
Face recognition from image sets acquired under unregulated and uncontrolled settings, such as at large distances, low resolutions, varying viewpoints, illumination, pose, and atmospheric conditions, is challenging. Face feature aggregation, which in
Externí odkaz:
http://arxiv.org/abs/2307.10237
Publikováno v:
2022 IEEE International Joint Conference on Biometrics (IJCB), Abu Dhabi, United Arab Emirates, 2022, pp. 1-9
Contactless fingerprint matching using smartphone cameras can alleviate major challenges of traditional fingerprint systems including hygienic acquisition, portability and presentation attacks. However, development of practical and robust contactless
Externí odkaz:
http://arxiv.org/abs/2307.05563
Autor:
Fedorishin, Dennis, Mohan, Deen Dayal, Jawade, Bhavin, Setlur, Srirangaraj, Govindaraju, Venu
Learning to localize the sound source in videos without explicit annotations is a novel area of audio-visual research. Existing work in this area focuses on creating attention maps to capture the correlation between the two modalities to localize the
Externí odkaz:
http://arxiv.org/abs/2211.03019
Autor:
Nedungadi, Prema, Ramesh, Maneesha, Govindaraju, Venu, Rao, Bhavani, Berbeglia, Paola, Raman, Raghu
Publikováno v:
In International Journal of Information Management August 2024 77
This paper describes a system by which Unmanned Aerial Vehicles (UAVs) can gather high-quality face images that can be used in biometric identification tasks. Success in face-based identification depends in large part on the image quality, and a majo
Externí odkaz:
http://arxiv.org/abs/2102.08542