• Home
  • Laval University
  • Department of Electrical Engineering and Computer Engineering
  • Jean-François Lalonde
Jean-François Lalonde

Jean-François Lalonde
Laval University | ULAVAL · Department of Electrical Engineering and Computer Engineering

Ph.D.

About

136
Publications
26,262
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
3,182
Citations
Introduction
Jean-François Lalonde is an Associate Professor at the Department of Electrical Engineering and Computer Engineering, Laval University. Jean-François does research in Computer Vision, Artificial Intelligence and Computer Graphics.

Publications

Publications (136)
Preprint
Full-text available
Adaptive instance normalization (AdaIN) has become the standard method for style injection: by re-normalizing features through scale-and-shift operations, it has found widespread use in style transfer, image generation, and image-to-image translation. In this work, we present a generalization of AdaIN which relies on the whitening and coloring tran...
Preprint
Full-text available
Scene inference under low-light is a challenging problem due to severe noise in the captured images. One way to reduce noise is to use longer exposure during the capture. However, in the presence of motion (scene or camera motion), longer exposures lead to motion blur, resulting in loss of image information. This creates a trade-off between these t...
Article
Full-text available
Plankton imaging systems supported by automated classification and analysis have improved ecologists' ability to observe aquatic ecosystems. Today, we are on the cusp of reliably tracking plankton populations with a suite of lab‐based and in situ tools, collecting imaging data at unprecedentedly fine spatial and temporal scales. But these data have...
Preprint
Full-text available
Deep generative models like StyleGAN hold the promise of semantic image editing: modifying images by their content, rather than their pixel values. Unfortunately, working with arbitrary images requires inverting the StyleGAN generator, which has remained challenging so far. Existing inversion approaches obtain promising yet imperfect results, havin...
Article
Full-text available
This paper describes eight imagery datasets including around 12000 images grouped in 1220 sets. The images were captured inside an architectural model aimed at exploring the impact of shading panels on photobiological lighting parameters. The architectural model represents a generic space at 1:10 scale with a single side fully glazing façade used t...
Preprint
We propose a method to extrapolate a 360{\deg} field of view from a single image that allows for user-controlled synthesis of the out-painted content. To do so, we propose improvements to an existing GAN-based in-painting architecture for out-painting panoramic image representation. Our method obtains state-of-the-art results and outperforms previo...
Preprint
Full-text available
In image classification, it is common practice to train deep networks to extract a single feature vector per input image. Few-shot classification methods also mostly follow this trend. In this work, we depart from this established direction and instead propose to extract sets of feature vectors for each image. We argue that a set-based representati...
Article
Lens design extrapolation (LDE) is a data-driven approach to optical design that aims to generate new optical systems inspired by reference designs. Here, we build on a deep learning-enabled LDE framework with the aim of generating a significant variety of microscope objective lenses (MOLs) that are similar in structure to the reference MOLs, but w...
Article
Full-text available
This paper presents a computational method for spatial visualization and probability evaluations of window view access in architecture based on human eyes' vision fields and biophilic recommendations. Window view access establishes occupants' visual connections to outdoors. Window view access has not, yet, been discussed in terms of the typical vis...
Poster
Full-text available
This research analyzes the impact of biophilic ambiance transition by means of coloured surfaces to respond to lighting needs of individuals who inhabit northern regions. Biophilic design aims to reduce the distance between humans and nature through architecture, a condition which is enhanced in northern latitudes due to their extreme climate condi...
Preprint
Full-text available
Most image-to-image translation methods require a large number of training images, which restricts their applicability. We instead propose ManiFest: a framework for few-shot image translation that learns a context-aware representation of a target domain from a few images only. To enforce feature consistency, our framework learns a style manifold be...
Poster
Full-text available
The present research aims to promote colour combination in interior architecture to develop light-responsive ambiances for northern regions. Humans are highly sensitive to light and colour, thus their application should be properly applied in extreme latitudes where light varies over a year. Latest discoveries have proved that colour application in...
Preprint
Full-text available
Most image-to-image translation methods focus on learning mappings across domains with the assumption that images share content (e.g., pose) but have their own domain-specific information known as style. When conditioned on a target image, such methods aim to extract the style of the target and combine it with the content of the source image. In th...
Article
Most modern commodity imaging systems we use directly for photography—or indirectly rely on for downstream applications—employ optical systems of multiple lenses that must balance deviations from perfect optics, manufacturing constraints, tolerances, cost, and footprint. Although optical designs often have complex interactions with downstream image...
Article
Full-text available
Northern building envelopes must provide efficient indoor-outdoor connections based on photobiological-psychological needs of occupants for positive relationships with the sub-Arctic nature, particularly daylighting and day/night cycles. Envelope configurations of Northern Canada’s buildings have not yet considered such requirements. Potentials of...
Poster
Les architectes ont pratiqué le confort et le bien-être des occupants dans des conditions climatiques variées. Sous les hautes latitudes, la satisfaction du confort des occupants est une tâche complexe pour les concepteurs en raison des conditions climatiques extrêmes. Cette recherche porte sur l'évaluation du confort et du bien-être afin de souten...
Article
Full-text available
Rain fills the atmosphere with water particles, which breaks the common assumption that light travels unaltered from the scene to the camera. While it is well-known that rain affects computer vision algorithms, quantifying its impact is difficult. In this context, we present a rain rendering pipeline that enables the systematic evaluation of common...
Article
We present a simple, highly modular deep neural network (DNN) framework to address the problem of automatically inferring lens design starting points tailored to the desired specifications. In contrast to previous work, our model can handle various and complex lens structures suitable for real-world problems such as Cooke Triplets or Double Gauss l...
Presentation
This research focuses on developing adaptive high-performance envelopes to enable biophilic healthy buildings in Arctic climates responding to occupants’ wellbeing needs for relationships with the outdoor nature and to energy efficiency requirements. Building envelopes are the main element connecting indoors to the extreme cold weather and drastic...
Preprint
Full-text available
We introduce Persistent Mixture Model (PMM) networks for representation learning in the few-shot image classification context. While previous methods represent classes with a single centroid or rely on post hoc clustering methods, our method learns a mixture model for each base class jointly with the data representation in an end-to-end manner. The...
Preprint
Full-text available
Much of the focus in the object detection literature has been on the problem of identifying the bounding box of a particular class of object in an image. Yet, in contexts such as robotics and augmented reality, it is often necessary to find a specific object instance---a unique toy or a custom industrial part for example---rather than a generic obj...
Chapter
Full-text available
Few-shot image classification aims at training a model from only a few examples for each of the “novel” classes. This paper proposes the idea of associative alignment for leveraging part of the base data by aligning the novel training instances to the closely related ones in the base training set. This expands the size of the effective novel traini...
Preprint
Recent work has demonstrated that deep learning approaches can successfully be used to recover accurate estimates of the spatially-varying BRDF (SVBRDF) of a surface from as little as a single image. Closer inspection reveals, however, that most approaches in the literature are trained purely on synthetic data, which, while diverse and realistic, i...
Preprint
Rain fills the atmosphere with water particles, which breaks the common assumption that light travels unaltered from the scene to the camera. While it is well-known that rain affects computer vision algorithms, quantifying its impact is difficult. In this context, we present a rain rendering pipeline that enables the systematic evaluation of common...
Conference Paper
This research aims at studying the potentials of single-skin and multi-skin envelopes with different window sizes to promote biophilic, healthy lighting and thermal performance of buildings in extreme sub-Arctic climatic conditions. Single-skin envelopes with low window-to-wall ratios (WWR) are most often recommended for sub-Arctic climates to redu...
Conference Paper
The aim of this research is to develop climate-based lighting adaptation scenarios for proper photobiological responses in high-performance biophilic buildings throughout the year. Photobiological responses refer to image forming (IF) and non-image forming (NIF) effects of light after reaching human eyes. IF effects enable vision. NIF effects regul...
Article
Modes of representation are key elements in communicating the properties of natural and built environments. This research proposes a capture and representation method for daylighting dynamics in interior and exterior spaces for the photopic (daytime vision) and melanopic (biological clock) portions of the electromagnetic spectrum. The proposed repr...
Preprint
Full-text available
Augmented reality devices require multiple sensors to perform various tasks such as localization and tracking. Currently, popular cameras are mostly frame-based (e.g. RGB and Depth) which impose a high data bandwidth and power usage. With the necessity for low power and more responsive augmented reality systems, using solely frame-based sensors imp...
Article
Full-text available
This study investigates shading panels' (SPs) impacts on daylighting features in a lab scale model in terms of parameters representing potential human eyes' biological responses identified as image forming (IF) and non-image forming (NIF). IF responses enable vision and NIF responses regulate internal body clocks known as circadian clocks. Human-ce...
Article
Full-text available
This paper develops an integrated design framework of adaptive building façades (ABFs) to respond to photobiological and thermal needs of occupants, biophilic factors, energy requirements and climatic features in Northern Canada, i.e. near and above 50°N. The paper discusses the importance of biophilic and photobiological factors and ABFs to improv...
Preprint
Full-text available
Computer vision datasets containing multiple modalities such as color, depth, and thermal properties are now commonly accessible and useful for solving a wide array of challenging tasks. However, deploying multi-sensor heads is not possible in many scenarios. As such many practical solutions tend to be based on simpler sensors, mostly for cost, sim...
Conference Paper
We developed a Convolutional Neural Network to estimate depth on wide-angle images using panomorph lens with controlled distortion. We simulated three different lens model and compared their performances based on their zone of augmented resolution.
Article
Photometric Stereo (PS) under outdoor illumination remains a challenging, ill-posed problem due to insufficient variability in illumination. Months-long capture sessions are typically used in this setup, with little success on shorter, single-day time intervals. In this paper, we investigate the solution of outdoor PS over a single day, under diffe...
Preprint
Full-text available
Few-shot image classification aims at training a model by using only a few (e.g., 5 or even 1) examples of novel classes. The established way of doing so is to rely on a larger set of base data for either pre-training a model, or for training in a meta-learning context. Unfortunately, these approaches often suffer from overfitting since the models...
Presentation
Full-text available
This research develops climate-responsive adaptation scenarios through which the indoor lighting ambiance could be adjusted to photobiological needs of occupants for different activities. Photobiological needs refer to image-forming (IF) and non-image forming (NIF) requirements of individuals to perform a particular activity. Recent photobiological...
Presentation
Full-text available
This research explores health and well-being responses of occupants to the daylighting design of northern latitude buildings. Photobiological and biophilic studies have shown the significant impacts of daylighting and natural cycles on occupants’ health and well-being. Northern people have forced to spend most of their time in buildings due to stro...
Poster
Full-text available
A smart portable device is proposed to monitor and analyze photobiological parameters of light, i.e. image forming and non-image forming, in the space and control artificial lighting and building façades systems with respect to occupants’ behavior and local photoperiods. The proposed device could operate in a real-time spatial basis to monitor inte...
Preprint
Full-text available
We present a method to estimate lighting from a single image of an indoor scene. Previous work has used an environment map representation that does not account for the localized nature of indoor lighting. Instead, we represent lighting as a set of discrete 3D lights with geometric and photometric parameters. We train a deep neural network to regres...
Article
We propose for the first time a deep learning approach in assisting lens designers to find a lens design starting point. Using machine learning, lens design databases can be expanded in a continuous way to produce high-quality starting points from various optical specifications. A deep neural network (DNN) is trained to reproduce known forms of des...
Poster
Full-text available
This study discusses the potential of climate-responsive building envelopes (CRBEs) to address biophilic and human-centric lighting factors inside buildings and satisfy occupants’ needs. Biophilic and human-centric lighting guidelines emphasize on image forming (IF) and non-image-forming (NIF) responses of building occupants to light and natural cy...
Poster
Full-text available
This research aims at developing shading panels of building envelopes to fulfill occupants’ health and well-being in terms of image forming (IF) and non-image forming (NIF) responses to light and natural cycles. In the last few years, photobiological studies have brought attention to the health and well-being of occupants in terms of IF and NIF res...
Poster
Full-text available
This research aims at developing biophilic adaptive façades as a promising solution to promote occupants’ health and building energy efficiency in the extreme-cold climate of Quebec's northern territories. Adaptive façades point to the (self-) adjustment of façade components (such as shading panels, windows and blinds) to occupants’ needs and envir...
Presentation
Full-text available
This research aims at developing biophilic adaptive façades as a promising solution to promote occupants’ health and building energy efficiency in the extreme-cold climate of Quebec's northern territories. Adaptive façades point to the (self-) adjustment of façade components (such as shading panels, windows and blinds) to occupants’ needs and envir...
Article
Nearly every commodity imaging system we directly interact with, or indirectly rely on, leverages power efficient, application-adjustable black-box hardware image signal processing (ISPs) units, running either in dedicated hardware blocks, or as proprietary software modules on programmable hardware. The configuration parameters of these black-box I...
Preprint
We present a neural network that predicts HDR outdoor illumination from a single LDR image. At the heart of our work is a method to accurately learn HDR lighting from LDR panoramas under any weather condition. We achieve this by training another CNN (on a combination of synthetic and real images) to take as input an LDR panorama, and regress the pa...
Preprint
Full-text available
We propose a real-time method to estimate spatiallyvarying indoor lighting from a single RGB image. Given an image and a 2D location in that image, our CNN estimates a 5th order spherical harmonic representation of the lighting at the given location in less than 20ms on a laptop mobile graphics card. While existing approaches estimate a single, glo...
Preprint
Relighting is an essential step in artificially transferring an object from one image into another environment. For example, a believable teleconference in Augmented Reality requires a portrait recorded in the source environment to be displayed and relit consistent with the light configuration of the destination scene. In this paper, we investigate...
Article
Full-text available
Large scenes such as building facades and other architectural constructions often contain repeating elements such as identical windows and brick patterns. In this paper, we present a novel approach that improves the resolution and geometry of 3D meshes of large scenes with such repeating elements. By leveraging structure from motion reconstruction...
Presentation
Full-text available
Le concours Ma thèse en 180 secondes est l'occasion pour les doctorants de présenter leur sujet de recherche en termes simples à un auditoire diversifié.
Preprint
We propose a data-driven learned sky model, which we use for outdoor lighting estimation from a single image. As no large-scale dataset of images and their corresponding ground truth illumination is readily available, we use complementary datasets to train our approach, combining the vast diversity of illumination conditions of SUN360 with the radi...
Article
Full-text available
This paper proposes the biophilic design approach as a plausible hypothesis for the challenging conditions related to living and working in extreme cold climates. Biophilic design has recently been developed to overcome the adverse effects of the built environment and to improve human well-being by redefining the human-nature relationship. Yet, bio...
Article
Predicting the short-term power output of a photovoltaic panel is an important task for the efficient management of smart grids. Short-term forecasting at the minute scale, also known as nowcasting, can benefit from sky images captured by regular cameras and installed close to the solar panel. However, estimating the weather conditions from these i...
Preprint
Predicting the short-term power output of a photovoltaic panel is an important task for the efficient management of smart grids. Short-term forecasting at the minute scale, also known as nowcasting, can benefit from sky images captured by regular cameras and installed close to the solar panel. However, estimating the weather conditions from these i...