Related papers: Pre-training with Random Orthogonal Projection Ima…
Unsupervised learning for monocular camera motion and 3D scene understanding has gained popularity over traditional methods, relying on epipolar geometry or non-linear optimization. Notably, deep learning can overcome many issues of…
Downlink reconfigurable intelligent surface (RIS)-assisted multi-input-multi-output (MIMO) systems are considered with far-field, near-field, and hybrid-far-near-field channels. According to the angular or distance information contained in…
Image-to-recipe retrieval is a challenging vision-to-language task of significant practical value. The main challenge of the task lies in the ultra-high redundancy in the long recipe and the large variation reflected in both food item…
Image metrics predict the perceived per-pixel difference between a reference image and its degraded (e. g., re-rendered) version. In several important applications, the reference image is not available and image metrics cannot be applied.…
Mask-based lensless imagers are smaller and lighter than traditional lensed cameras. In these imagers, the sensor does not directly record an image of the scene; rather, a computational algorithm reconstructs it. Typically, mask-based…
Recent advancements in masked image modeling (MIM) have made it a prevailing framework for self-supervised visual representation learning. The MIM pretrained models, like most deep neural network methods, remain vulnerable to adversarial…
Self-supervised learning has made unsupervised pretraining relevant again for difficult computer vision tasks. The most effective self-supervised methods involve prediction tasks based on features extracted from diverse views of the data.…
Hyperspectral imaging is an advanced technique for precisely identifying and analyzing materials or objects. However, its integration with robotic grasping systems has so far been explored due to the deployment complexities and prohibitive…
Traditional AI-planning methods for task planning in robotics require a symbolically encoded domain description. While powerful in well-defined scenarios, as well as human-interpretable, setting this up requires substantial effort.…
Unsupervised semantic segmentation aims to automatically partition images into semantically meaningful regions by identifying global semantic categories within an image corpus without any form of annotation. Building upon recent advances in…
Undersampling is a common method in Magnetic Resonance Imaging (MRI) to subsample the number of data points in k-space, reducing acquisition times at the cost of decreased image quality. A popular approach is to employ undersampling…
Recently, masked image modeling (MIM) has offered a new methodology of self-supervised pre-training of vision transformers. A key idea of efficient implementation is to discard the masked image patches (or tokens) throughout the target…
Several recent works have directly extended the image masked autoencoder (MAE) with random masking into video domain, achieving promising results. However, unlike images, both spatial and temporal information are important for video…
Human activity recognition (HAR) with deep learning models relies on large amounts of labeled data, often challenging to obtain due to associated cost, time, and labor. Self-supervised learning (SSL) has emerged as an effective approach to…
Path representations are critical in a variety of transportation applications, such as estimating path ranking in path recommendation systems and estimating path travel time in navigation systems. Existing studies often learn task-specific…
SOM is a type of unsupervised learning where the goal is to discover some underlying structure of the data. In this paper, a new extraction method based on the main idea of Concurrent Self-Organizing Maps (CSOM), representing a…
Linear reduced-order modeling (ROM) simplifies complex simulations by approximating the behavior of a system using a simplified kinematic representation. Typically, ROM is trained on input simulations created with a specific spatial…
Recent progress in human shape learning, shows that neural implicit models are effective in generating 3D human surfaces from limited number of views, and even from a single RGB image. However, existing monocular approaches still struggle…
Rotoscoping, the detailed delineation of scene elements through a video shot, is a painstaking task of tremendous importance in professional post-production pipelines. While pixel-wise segmentation techniques can help for this task,…
Instruction-based image editing (IIE) models have recently demonstrated strong capability in modifying specific image regions according to natural language instructions, which implicitly requires identifying where an edit should be applied.…