Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 20 of 102 for “"Cross-attention"”.
-
Image Geo-localization with Cross-Attention
… a new method under this framework which uses cross-attention between the query image and a database of satellite imagery with known geotags. Our experiments fit within three broad categories: 1) evaluating the ability of image localization approaches to generalize to unseen regions; 2) …
-
Hierarchical Deep Reinforcement Learning with Cross-Attention and Planning for Autonomous Roundabout Navigation
Autonomous vehicle control is an important sub-field of autonomous vehicle research. Many challenges remain to improve the safety and performance of autonomous vehicle control systems in urban driving environments. One such urban driving environment is the roundabout junction, which presents its …
-
Correlation-Aware Variational Autoencoder with Cross-Attention for Zero-Day Network Anomaly Detection
… Gaussian posteriors and a Dual-Directional Cross-Attention mechanism that verifies semantic consistency between packet headers and payloads. Extensive evaluation on the CICIDS-2017 and CICIDS-2018 datasets demonstrates CAVAD’s superior performance compared to state-of-the-art baselines. On …
-
A BERT-ResNet Cross-Attention Fusion Network and Modality Utilization Assessment for Multimodal Sentiment Classification
… that can accurately interpret sentiment across modalities. We implement a model using BERT for textual features and ResNet for visual features. A cross-attention fusion module aligns the modalities for joint representation. We conduct experiments on the MVSA-Single and MVSA-Multiple …
-
CMTN – Cross-Modal Triage Network: A Multimodal Deep Learning Framework for Severity-Based Triage and Visual Explainability in Chest Radiographs
… behind routine ones. This thesis introduces the Cross-Modal Triage Network (CMTN), a multimodal framework fusing a Swin Transformer V2 visual backbone with a PubMedBERT textual encoder via Gated Cross-Attention to prioritize chest radiographs by severity while jointly predicting fourteen …
-
SwinFSR: Stereo Image Super-Resolution using SwinIR and Frequency Domain Knowledge
… (stereoSR) has attracted significant attention in recent years due to the extensive deployment of dual cameras in mobile phones, autonomous vehicles and robots. In this work, we propose a new StereoSR method, named SwinFSR, based on an extension of SwinIR, originally designed for …
-
Towards Robust Person Re-Identification: Learning Invariant Features Under Clothing Changes and Occlusions
… feature representations capable of generalizing across these challenging scenarios. To tackle Cloth-Changing Re-ID (CCRe-ID), we first introduce a multi-modality framework that incorporates contrastive viewpoint-aware shape learning, enhancing shape robustness against viewpoint variations. Next, …
-
Deep Brain Dynamics and Images Mining for Tumor Detection and Precision Medicine
… In this thesis, we propose a novel deep cross-attention learning algorithm that maximizes the subtle dynamics mining from each of the input modalities and then boosts feature fusion capability. More specifically, we have designed a Multimodal Cross-Attention Module (MM-CAM), equipped with …
-
SD Map Localization: A Deep Learning Approach
… or Transformers, a Map Matching Network based on cross-attention, and a Point Prediction Network consisting of a simple Multi-Layer Perceptron. The model encodes ego trajectories and map links, matches the map data with the vehicle’s trajectory, and predicts precise location. Our results show that …
-
Hyperbolic vision–language embeddings and loss functions for multimodal meme classification
… Pre-training) encoders with hyperbolic gated cross-attention layers. We adopt Flamingo's frozen-encoder design because the benchmark requires rapid iteration, and lightweight adapters allow efficient experimentation across different geometric configurations. Initial experiments revealed a core …
-
REPRESENTATION LEARNING FOR VISUAL TASKS: A STUDY OF ATTENTION AND INFORMATION SELECTION
… visual representation learning by optimizing attention mechanisms and information selection strategies within deep learning models. Standard approaches often process images independently and compress them into single global descriptors, limiting performance on tasks requiring contextual …
-
Enhancing Argument Mining with Interpretable Sources and Semantic Information
… a RoBERTa-based Siamese model with multi-head cross attention to perform fine-grained and context-sensitive ARC. Event Knowledge Alignment for Argument Relation Classification proposes the EKA-ARC framework, which enriches argument components by aligning them with structured event knowledge …
-
Biometric Leakage from Generative Models and Adversarial Iris Swapping for Spoofing Eye-based Authentication
… images. Next, we evaluate the effectiveness of Cross-Attention GANs for iris-swapping attacks, demonstrating their ability to enable presentation attacks that spoof iris-recognition systems. Our experiments across several standard iris and VR datasets have an attack success rate of 100% within …
-
Unified driving maneuver detection and classification from monocular video
… preserves the motion-sensitive signals that cross-attention tends to dilute. For temporal action detection, we use a four-stage pipeline: Mamba-based causal sequence processing, boundary-aware localization, proposal generation, and graph convolutional refinement. The selection mechanism of …
-
Advancing Temporal Action Localization: Efficient Large Model Adaptation and Open-Vocabulary Recognition in Videos
… untrimmed videos, modifying intermediate layers across various temporal ranges to enhance video features. It enables end-to-end adaptation of billion-parameter models like VideoMAEv2. The OVFormer framework addresses Open-Vocabulary TAL by generating rich class descriptions using a language model, …
-
TOWARDS A UNIFIED SELF-SUPERVISED LEARNING FRAMEWORK FOR DATA FUSION IN REMOTE SENSING
… framework based on masked autoencoders that uses cross-attention to perform early and feature-level data fusion between synthetic aperture radar and multispectral optical data - two modalities with a significant domain gap. Our empirical findings demonstrate that Fus-MAE can effectively compete …
-
A Malware Classification Method Based on the Multi-Layer Feature Fusion of Malware Image Representations and Opcode Markov Images
… model introduces two feature fusion modules: a cross-attention-based fusion module and a multiplication-based low-level feature fusion module. These modules achieve effective fusion of deep features, thereby improving the accuracy of malware classification. Experimental results show that the …
-
Enhancing Emotion Recognition in Textual Conversation by Leveraging Common-sense
… and second, explicit incorporation by applying cross-attention on common-sense knowledge for each utterance. I demonstrate that the proposed methods perform well on IEMOCAP, a widely used dyadic conversation dataset with human-annotated emotions.
-
Self-supervised Pretraining of Vision Transformers for Earth Observation
… lacks foundation models that can be leveraged across societally impactful applications. In this thesis, I develop foundation models for Earth Observation by pretraining vision transformers (ViTs) using self-supervised learning. Using masked autoencoding, I develop one of the first ViTs …
-
Commonsense for Zero-Shot Natural Language Video Localization
… a knowledge graph, conditioned on the video, and cross-attention mechanisms to enhance the encoded video and pseudo-query vectors prior to localization. Through empirical evaluations on two benchmark datasets, we demonstrate that our model surpasses both zero-shot and weakly supervised baselines. …
Page 1 of 6