One-Shot Imitation from Observing Humans via Domain-Adaptive Meta-Learning
Learn domain-adaptive meta-learning for one-shot imitation, enabling robots to quickly learn new tasks by observing human demonstrations and adapting across domains.
A Sparsely Labeled Dataset for Action Classification and Localization - SLAC
Explore a novel dataset for action classification and localization, covering data collection, training, validation, and evaluation methods for human action detection.
Pixel Recurrent Neural Networks
Explore cutting-edge image generation techniques, focusing on Pixel Recurrent Neural Networks. Learn about model architectures, RNN concepts, and LSTM equations for advanced AI applications.
Deep Reinforcement Learning for Object Tracking
Explore action-driven object tracking using deep reinforcement learning, covering policy, value functions, and training approaches for improved visual tracking performance.
High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs
Explore high-resolution image synthesis and semantic manipulation using conditional GANs, covering advanced techniques like coarse-to-fine generation and multi-scale discriminators for improved photorealism.
What If We Do Not Have Multiple Videos of the Same Action - Video Action Localization
Explore video action localization techniques for scenarios with limited action samples, covering key challenges, innovative approaches, and experimental results.
Global Data Association for Multiple Pedestrian Tracking
Explore advanced data association techniques for multi-target tracking, from GMCP to GMMCP, and a novel approach for crowd tracking using Binary Quadratic Programming and Modified Frank-Wolfe algorithm.
Fundamental Matrix and Stereo Vision - Lecture 13
Explore fundamental matrices, stereo photosynthesis, and random sampling consensus in image processing and computer vision applications.
Filtering in Computer Vision - Lecture 2
Learn about image filtering techniques, including noise reduction, derivatives, convolution, and Gaussian filters. Explore practical applications in computer vision using MATLAB.
Interest Point Detection in Computer Vision - Lecture 4
Explore interest point detection techniques, including Harris Corner Detector and auto-correlation, with applications in computer vision and image processing.
Pyramids in Image Processing - Lecture 5
Explore image pyramids, including Gaussian and Laplacian types, their applications in image processing, and techniques like reduction, expansion, and compression.
Facial Recognition: Eigenvectors and Covariance Matrices - Lecture 14
Explore facial recognition technology, from simple approaches to advanced techniques using eigenvectors and covariance matrices. Learn about challenges and solutions in this field.
Bag of Words Model for Image Classification - Lecture 16
Explore bag-of-features for image classification, including feature extraction, quantization, and spatial pyramid matching. Learn about datasets and applications in scene and action recognition.
Hough Transform for Image Feature Extraction - Lecture 17
Explore image feature extraction, line fitting techniques, and the Hough Transform for shape detection in computer vision, including practical applications and challenges.
Karhunen-Loève Transform (KLT) Algorithm for Tracking - Lecture 10
Explore tracking techniques, including KLT algorithm, for computer vision applications. Learn about translation models, Jacobian matrices, and multi-camera setups.
Scale-Invariant Feature Transform (SIFT) - Lecture 6
Explore SIFT algorithm's principles, advantages, and applications in computer vision, including scale-space representation and feature detection techniques.
Visual Analysis of Extremely Dense Crowded Scenes
Analyzing dense crowd scenes using computer vision techniques for counting, detecting, and tracking individuals. Explores multiple data sources, context, and neighborhood motion to address challenges in extremely crowded environments.
Human Detection, Tracking and Segmentation in Surveillance Video
Explore advanced techniques for human detection, tracking, and segmentation in surveillance videos, focusing on scene-specific learning and occlusion handling for improved accuracy and efficiency.
Action Recognition in Temporarily Untrimmed Videos
Action recognition in untrimmed videos: challenges, datasets, and annotation techniques for improving accuracy in real-world scenarios with ambiguous content and varying video qualities.
Statistical and Spatial Consensus Collection for Detector Adaption
Explore detector adaptation techniques for improved pedestrian detection across domains, focusing on statistical and spatial consensus methods to enhance accuracy and generalization.
Visual Geo-Localization and Location-Aware Image Understanding
Explore visual geo-localization techniques, including image matching, geospatial pruning, and segment localization. Learn about business recognition and multi-hypotheses character recognition for location-aware image understanding.