• IEEE Transactions of Neural Networks and Learning Systems
• IEEE Signal Processing Letters
• IEEE Transactions on Information Forensics and Security
• IEEE Transactions on Image Processing
• Springer International Journal of Computer Vision
• ICCV'19-23
• CVPR'20-24
• ECCV'20-24
• AAAI-2022
• NeurIPS 2021-2023

Decorum enables users to control the scene generation process with natural language by adopting language-based representations at each stage. This enables us to harness recent advancements in Large Language Models (LLMs) to model language-to-language mappings.

TalkinNeRF is a unified NeRF-based network that represents the holistic 4D human motion. Given a monocular video of a subject, we learn corresponding modules for the body, face, and hands, that are combined together to generate the final result.

AVFace incorporates both modalities (audio and video) and accurately reconstructs the 4D facial and lip motion of any speaker, without requiring any 3D ground truth for training.

It is challenging to make the best use of multiple exemplars: the quality and alignment of each exemplar cannot be guaranteed. Using low-quality and mismatched images as references will impair the output results. To overcome these issues, we propose the efficient Headshot Image Super-Resolution with Multiple Exemplars network (HIME) method.

We propose Eye Authentication with PAD (EyePAD), a distillation-based method that trains a single network for EA and PAD while reducing the effect of forgetting.

To restore videos on recipient edge devices in real-time, we introduce an efficient video restoration network, EVRNet. EVRNet efficiently allocates parameters inside the network using alignment, differential, and fusion modules.

RAE uses hierarchical operational models to perform tasks in dynamically changing environments.

A semisupervised method for jointly extracting facial landmarks and restoring the degraded images by exploiting the semantic information from the landmarks.

Predicting landmarks directly on low resolution images is more effective than the current practice of aligning images after rescaling or super-resolution.

We leverage an attention-based model which learns to focus on different parts of a vehicle by conditioning the feature maps on visible key-points. We use triplet embedding to reduce the dimensionality of the features obtained from the ensemble of networks trained using different datasets.

In AAVER, the global appearance path captures macroscopic vehicle features while the orientation conditioned part appearance path learns to capture localized discriminative features by focusing attention on the most informative key-points.

We propose a simple two-stage algorithm to approximate the transformation. Images are first rectified to restore affine properties, then the scaling factor is compensated for each scene.

Following a Bayesian formulation, we disentangle the 3D pose of a face image explicitly by conditioning the landmark estimation on pose, making it different from multi-tasking approaches.

We present a novel architecture called H-CNN (Heatmap-CNN) acting on an N-dimensional input image which captures informative structured global and local features and thus favors accurate keypoint detecion in in-the wild face images.

Although a simple feed forward neural network can learn the mapping between input and output spaces, it cannot learn the inherent structural dependencies. We present a novel architecture called H-CNN (Heatmap-CNN) which captures structured global and local features and thus favors accurate keypoint detecion.

First Deep Learning based face verification system.

A coarse-to-fine approach for estimating the apparent age from unconstrained face images using deep convolutional neural networks (DCNNs). Also one of the first models in the Deep learning era

A brief history of developments in computer vision and artificial neural networks over the last forty years for the problem of image-based recognition. We then present the design details of a deep learning system for endto-end unconstrained face verification/recognition.

The end-to-end system consists of three modules for face detection, alignment and verification and is evaluated using the newly released IARPA Janus Benchmark A (IJB-A) dataset.

The proposed approach exploits two insights: (1) Features obtained from DCNN trained for face-identification task can be used for age estimation. (2) The three-layer neural network regression method trained on Gaussian loss performs better than traditional regression methods for apparent age estimation.

Local Deep Descriptor Regression (LDDR) is able to localize face landmarks of varying sizes, poses and occlusions with high accuracy. Deep Descriptors presented in this paper are able to uniquely and efficiently describe every pixel in the image and therefore can potentially replace traditional descriptors such as SIFT and HOG.