no code implementations • 20 Jan 2024 • Navin Ranjan, Andreas Savakis
To validate and assess our approach, we employ LRP-QViT across ViT, DeiT, and Swin transformer models on various datasets.
no code implementations • 5 Dec 2023 • Chowdhury Sadman Jahan, Andreas Savakis
Source-free OSDA (SF-OSDA) techniques eliminate the need to access source domain samples, but current SF-OSDA methods utilize only the known classes in the target domain for adaptation, and require access to the entire target domain even during inference after adaptation, to make the distinction between known and unknown samples.
1 code implementation • 2 Aug 2023 • Chowdhury Sadman Jahan, Andreas Savakis
We synthesize two such gradually worsening weather conditions on real images from two existing aerial imagery datasets, generating a total of four benchmark datasets.
no code implementations • 2 Aug 2023 • Chowdhury Sadman Jahan, Andreas Savakis
Addressing the rising concerns of privacy and security, domain adaptation in the dark aims to adapt a black-box source trained model to an unlabeled target domain without access to any source data or source model parameters.
no code implementations • 28 May 2022 • Alexander Avery, Andreas Savakis
Optical flow prediction stabilizes the training process, and enforces the learning of features that are relevant to the task of pose estimation.
1 code implementation • 20 Dec 2021 • Bruno Artacho, Andreas Savakis
We propose BAPose, a novel bottom-up approach that achieves state-of-the-art results for multi-person pose estimation.
Ranked #5 on Multi-Person Pose Estimation on MS COCO
no code implementations • 13 May 2021 • Nilesh Pandey, Andreas Savakis
Recovering badly damaged face images is a useful yet challenging task, especially in extreme cases where the masked or damaged region is very large.
no code implementations • 16 Apr 2021 • Navya Nagananda, Breton Minnehan, Andreas Savakis
Linear Discriminant Analysis (LDA) is commonly used for dimensionality reduction in pattern recognition and statistics.
no code implementations • 8 Apr 2021 • Abu Md Niamul Taufique, Andreas Savakis, Michael Braun, Daniel Kubacki, Ethan Dell, Lei Qian, Sean M. O'Rourke
Siamese deep-network trackers have received significant attention in recent years due to their real-time speed and state-of-the-art performance.
no code implementations • 24 Mar 2021 • Abu Md Niamul Taufique, Breton Minnehan, Andreas Savakis
In recent years, deep learning-based visual object trackers have achieved state-of-the-art performance on several visual object tracking benchmarks.
no code implementations • 20 Mar 2021 • Abu Md Niamul Taufique, Andreas Savakis, Jonathan Leckenby
Motion is computed and compared between the left and the right parts of each region of interest to estimate the symmetry score.
no code implementations • 20 Mar 2021 • Abu Md Niamul Taufique, Navya Nagananda, Andreas Savakis
Synthetic Aperture Radar (SAR) imagery has diverse applications in land and marine surveillance.
no code implementations • 19 Mar 2021 • Abu Md Niamul Taufique, Chowdhury Sadman Jahan, Andreas Savakis
Our results on three popular DA datasets demonstrate that our method outperforms many existing state-of-the-art DA methods with access to the entire target domain during adaptation.
1 code implementation • 18 Mar 2021 • Bruno Artacho, Andreas Savakis
We propose OmniPose, a single-pass, end-to-end trainable framework, that achieves state-of-the-art results for multi-person pose estimation.
Ranked #1 on Pose Estimation on MS COCO
no code implementations • 29 Nov 2020 • Abu Md Niamul Taufique, Andreas Savakis
Vehicle re-identification is an important computer vision task where the objective is to identify a specific vehicle among a set of vehicles seen at various viewpoints.
2 code implementations • CVPR 2020 • Bruno Artacho, Andreas Savakis
Our results on multiple datasets demonstrate that UniPose, with a ResNet backbone and Waterfall module, is a robust and efficient architecture for pose estimation obtaining state-of-the-art results in single person pose detection for both single images and videos.
Ranked #2 on Pose Estimation on UPenn Action
1 code implementation • 6 Dec 2019 • Bruno Artacho, Andreas Savakis
We propose a new efficient architecture for semantic segmentation, based on a "Waterfall" Atrous Spatial Pooling architecture, that achieves a considerable accuracy increase while decreasing the number of network parameters and memory footprint.
Ranked #11 on Semantic Segmentation on PASCAL VOC 2012 val
1 code implementation • 5 Sep 2019 • Nilesh Pandey, Andreas Savakis
We present Poly-GAN, a novel conditional GAN architecture that is motivated by Fashion Synthesis, an application where garments are automatically placed on images of human models at an arbitrary pose.
Ranked #2 on Virtual Try-on on Deep-Fashion
no code implementations • CVPR 2019 • Breton Minnehan, Andreas Savakis
We propose a data-driven approach for deep convolutional neural network compression that achieves high accuracy with high throughput and low memory requirements.
no code implementations • 27 Sep 2018 • Shagan Sah, Dheeraj Peri, Ameya Shringi, Chi Zhang, Miguel Dominguez, Andreas Savakis, Ray Ptucha
Along with MMVR, we propose two improvements to the text conditioned image generation.
no code implementations • 20 Jun 2018 • Breton Minnehan, Andreas Savakis
We propose a novel technique for training deep networks with the objective of obtaining feature representations that exist in a Euclidean space and exhibit strong clustering behavior.
no code implementations • 1 Dec 2016 • Jefferson Ryan Medel, Andreas Savakis
Automating the detection of anomalous events within long video sequences is challenging due to the ambiguity of how such events are defined.