Work for a Member company and need a Member Portal account? Register here with your company email address.
Copyright
Camera Culture
A City-Stack Powered by Decentralized AIOverviewMIT Kumbhathon, hosted at the MIT Media Lab, is pioneering a revolutionary approach t…
The acquisition of training data is crucial for machine learning applications. Data markets can increase the supply of data, particularly i…
As AI evolves beyond screen assistants and into dimensional applications, decentralization emerges as the critical factor for unlocking i…
The challenge of understanding pandemics, illustrated by the recent COVID-19 outbreak, has significantly impacted global health and economi…
Press: - MIT News: Using reflections to see the world from new points of view- TechXplore, EuropaPress, La Nacion …
Cloud based machine learning inference is an emerging paradigm where users query with their data to a service provider who runs a ML model …
The goal of non-line-of-sight (NLOS) imaging is to image objects that lie outside the camera’s field of view. We use two-bounce light to pe…
Data is the main fuel of the modern world, enabling artificial intelligence and driving technological growth. The demand for data has grown…
Computer Vision applications are making a remarkable impact on society and advancing progress in Machine Learning. Our project aims to desi…
Although lidar is widely used for mapping the 3D geometry of surfaces, the technology has historically been challenged by specular, or mirr…
Point clouds are an increasingly ubiquitous input modality and the raw signal can be efficiently processed with recent progress in deep lea…
Image-based virtual try-on involves synthesizing perceptually convincing images of a model wearing a particular garment and has garnered si…
By predicting sparse shadow cues, our physics-inspired machine learning algorithm can reconstruct the underlying 3D scene. Abstract. W…
We tackle the question of how to benchmark reconstruction of inputs from deep neural networks representations. This inverse problem is of g…
Differential Privacy offers strong guarantees such as immutable privacy under any post-processing. In this work, we propose a differentiall…
Classes of set functions along with a choice of ground set are a bedrock to determine and develop corresponding variants of greedy algorith…
We introduce a differentially private method to measure nonlinear correlations between sensitive data hosted across two entities. We provid…
Wireless channels can be inherently privacy-preserving by distorting the received signals due to channel noise, and superpositioning multip…
We propose sanitizer, a framework for secure and task-agnostic data release. While releasing datasets continues to make a big impact in var…
Abstract:We introduce DeepABM, a framework for agent-based modeling that leverages geometric message passing of graph neural networks for s…
Recent deep learning models have shown remarkable performance in image classification. While these deep learning systems are getting closer…
In this work, we introduce FedML, an open research library and benchmark that facilitates the development of new 'federated learning algori…
NoPeek-Infer: Preventing face reconstruction attacks in distributed inference after on-premise training (Won FG-2021 Mukh Best Paper Runner…
We propose an improved private count-mean-sketch data structure and show its applicability to differentially private contact trac…
We present a stochastic scheme for splitting the client data into privatized shares that are transmitted to the server in such settings. Th…
Papers from SafePaths: https://github.com/PrivateKit/PrivacyDocumentsSafe Paths is an MIT-led, free, open source technology that …
Split Learning: Distributed deep learning without sharing raw data Project Page: https://splitlearning.github.io/Abstract: C…
Rethinking photography optics in the time dimensionWhat if we could design optics in time instead of space?
Seeing through dense, dynamic, and heterogeneous fog conditions. The technique, based on visible light, uses hardware that is similar to LI…
Time of Flight 3D cameras like the Microsoft Kinect are prevalent in computer vision and computer graphics. In such devices, the power of a…
Object Classification through Scattering Media with Deep LearningA method for classifying objects hidden behind a scattering layer wit…
We introduce MetaQNN, a meta-modeling algorithm based on reinforcement learning to automatically generate high-performing CNN architectures…
Lensless imaging with compressive ultrafast sensingTraditional cameras require a lens and a mega-pixel sensor to capture images. The lens f…
Computation photography applied to imaging the middle ear for infections.
We demonstrate a smartphone based spectrometer design that is standalone and supported on a wireless platform. The device is inherently low…
Terahertz time-gated spectral imaging for content extraction through layered structures.
How to see through tissueWe demonstrate a new method to image through scattering materials like tissue and fog. The demonstration includes …
In this visual brainstorming, we present the next 30 years of VR in a set of concept designs.
Skin and tissue perfusion measurements are important parameters for diagnosis of wounds and burns, and for monitoring plastic and reconstru…
Our deformable camera exploits new, flexible form factors for imaging in turbid media. In this study we enable a brush-like form factor wit…
We believe that tough global health problems require an innovation pipeline. We must bring together the people and providers facing health …
Fluorescence lifetime imaging is a significant bio-imaging tool that finds important applications in life-sciences. Widely known applicatio…
Eye exams via a slit lamp are critical in early diagnosis of diseases such as cataracts, corneal injury, and pterygia, in order to avert vi…
This concept gallery shows the chain of startups and ideas that will follow after the emergence of self-driving cars.
A new method to detect and distinguish between different types of fluorescent materials. The suggested technique has provided a dramaticall…
Using a femtosecond laser and a camera with a time resolution of about one trillion frames per second, we recover objects hidden out of sig…
Our architecture takes a hybrid approach to microwaves and treats them like waves of light. Most other work places antennas in a 2D arrange…
The ability to record images with extreme temporal resolution enables a diverse range of applications, such as time-of-flight depth imaging…
Asthma is the most common chronic illness among children. The skills required to diagnose it make it an even greater concern. Our solution …
We present a novel framework to extend the dynamic range of images called Unbounded High Dynamic Range (UHDR) photography with a modulo cam…
Locating and classifying florescent tags behind turbid layers using time-resovled inversion Using time resolved and sparse optimizatio…
We are exploring mathematical modeling of time-of-flight imaging problems and solutions.
Time-of-flight (ToF) cameras are commercialized consumer cameras that provide a depth map of a scene, with many applications in computer vi…
The major challenge in preventing blindness is identifying patients and bringing them to specialty care. Diseases that affect the retina, t…
We have developed a camera system that captures movies at an effective rate of approximately one trillion frames per second. In one frame o…
Can a person look at a portable display, click on a few buttons, and recover his or her refractive condition? Our optometry solution combin…
We use time-resolved information in an iterative optimization algorithm to recover reflectance of a three-dimensional scene hidden behind a…
We demonstrate a new technique that allows a camera to rapidly acquire reflectance properties of objects "in the wild" from a single viewpo…