Senior Lecturer
Computer Vision, Deep Learning
Publications of specific relevance to the Centre for Multimodal AI
Compress & Cache: Vision token compression for
efficient generation and retrievalFAM Diffusion: Frequency and Attention Modulation for High-Resolution Image Generation with Stable DiffusionVladVA: Discriminative Fine-tuning of LVLMsEdge-SD-SR: Low Latency and Parameter Efficient On-device Super-Resolution with Stable Diffusion via Bidirectional ConditioningVision-Free Retrieval: Rethinking Multimodal Search with Textual Scene DescriptionsMemsvd: Long-Range Temporal Structure Capturing Using Incremental SVDMobileQuant: Mobile-friendly Quantization for On-device Language ModelsEfficient Unsupervised Visual Representation Learning with Explicit Cluster BalancingEfficient Unsupervised Visual Representation Learning with Explicit Cluster BalancingQBB: Quantization with Binary Bases for LLMsCemiFace: Center-based Semi-hard Synthetic Face Generation for Face RecognitionEfficient Vision-Language pre-training via domain-specific learning for human activitiesLanguage-Aware Soft Prompting: Text-to-Text Optimization for Few- and Zero-Shot Adaptation of V &L ModelsBlack Box Few-Shot Adaptation for Vision-Language modelsFS-DETR: Few-Shot DEtection TRansformer with prompting and without re-trainingReGen: A good Generative zero-shot video classifier should be RewardedHyperReenact: One-Shot Reenactment via Jointly Learning to Refine and Retarget FacesBayesian Prompt Learning for Image-Language Model GeneralizationLASP: Text-to-Text Optimization for Language-Aware Soft Prompting of Vision & Language ModelsFrom Keypoints to Object Landmarks via Self-Training Correspondence: A Novel Approach to Unsupervised Landmark DiscoveryPart-based Face Recognition with Vision
TransformersFinding Directions in GAN’s Latent Space for
Neural Face ReenactmentEdgeViTs: Competing Light-Weight CNNs on Mobile Devices with Vision TransformersPre-training Strategies and Datasets for Facial Representation LearningSpace-time Mixing Attention for Video TransformerWarpedGANSpace: Finding non-linear RBF paths in GAN latent spaceBit-Mixer: Mixed-precision networks with runtime bit-width selectionAffective Processes: stochastic modelling of temporal context for emotion and facial expression recognitionKnowledge distillation via softmax regression representation learningHigh-Capacity Expert Binary NetworksSelf-supervised Learning of Person-specific Facial Dynamics for Automatic Personality RecognitionA Transfer Learning approach to Heatmap Regression for Action Unit intensity estimationUnsupervised Learning of Object Landmarks via Self-Training CorrespondenceAnimaWeb: A Large-Scale Hierarchical Dataset of Annotated Animal FacesFAN-Face: a Simple Orthogonal Improvement to Deep Face RecognitionBATS: Binary ArchitecTure SearchT-Net: Parametrizing Fully Convolutional Nets with a Single High-Order TensorHierarchical Binary CNNs for Landmark Localization with Limited ResourcesLarge Pose 3D Face Reconstruction from a Single Image via Direct Volumetric CNN RegressionHow Far are We from Solving the 2D & 3D Face Alignment Problem? (and a Dataset of 230,000 3D Facial Landmarks)A Functional Regression Approach to Facial Landmark TrackingConvolutional aggregation of local evidence for large pose face alignmentHuman Pose Estimation via Convolutional Part Heatmap RegressionGauss-Newton Deformable Part Models for Face Alignment in-the-WildA Simple Baseline for Knowledge-Based Visual Question AnsweringGrants of specific relevance to the Centre for Multimodal AIReliable AI and Data OptimisationOn 2-7 December, several CMAI researchers will participate at the 39th Annual Conference on Neural Information Processing Systems (NeurIPS 2025), taking place in San Diego. NeurIPS is a prestigious annual academic conference and non-profit foundation that fosters the exchange of research in artificial intelligence (AI), machine learning (ML), and computational neuroscience. ... [more]
On 30 June - 5 July 2025, CMAI researchers will participate at the IEEE International Joint Conference on Neural Networks (IJCNN 2025), the flagship conference of the IEEE Computational Intelligence Society and the International Neural Network Society. The Centre for Multimodal AI will have a strong presence at the conference. The following papers authored/... [more]
