Alexandre Alapetite

interesting theory being proposed, and apparently this has a lot to do with the geometry and spacing of axons within myelin sheaths. a very low energy way to transmit data... and I guess this overcomes issued with system resistance. and speed. very cool, if indeed this is the way it works. https://lnkd.in/giqy9DtD

Wind speed? Measured by drones? Without any airspeed sensor? Yes, you read that right! In our latest work, we’ve managed to estimate wind velocity using only the drones (3 of them) themselves—no extra hardware, no expensive calibration needed. It’s not magic, it’s science! And I’m pretty sure that Gabriel Laupré, Pasquale Longobardi, and Jan Skaloud, who worked on this with me, would back me up here. 😉 Curious about the details? Dive into our paper to see how we did it (spoiler: there’s no voodoo involved). Paper: https://lnkd.in/gChBqYDZ

35

Here is an article covering my contribution to #safetronic2024 as an kind of summery: https://lnkd.in/eJhZ-c-f Thanks #FraunhoferIKS for the opportunity to write it and publishing :)

21

1 Comment

VR digital twin of office space with computer vision-based estimation of room occupancy and power consumption Paper Link: https://lnkd.in/gm4JVWvt Video Link: https://lnkd.in/g_k725Ux In the past years, energy consumption has increased rapidly due to many factors, including the rise in technology adoption. This has many downfalls, from higher costs to CO emissions. Human activities in offices and houses represent a considerable amount of energy usage. A digital twin (DT) of an open-plan common space is created, serving the purpose of remote room occupancy monitoring and automatic detection of energy consumption. A virtual reality (VR) model is developed and integrated to temperature, humidity and imaging sensors. For maintaining privacy, images are processed in local computers to measure occupancy levels and live video feed were never transmitted. The same set of imaging sensors were also used in a bespoke computer vision module for energy consumption estimation. The human avatars were mapped with high correlation with actual positions on floor. Our energy consumption algorithm accuracy obtained true positive rate of 92% and F1 score of 82%. Finally, all this information is transmitted and visualized to the 3D digital twin for remote monitoring and simulation. Abhishek Mukhopadhyay Naveen Talwar Himanshu Vishwakarma Rajshekar Reddy Shakti Srivastava Anasol Peña-Rios #vr #digitaltwin #bt #sdg7 #computervision #cv #virtualreality

56

Following our trilogy on disability-centered algorithms and policy, as a scientific evaluator I contributed to the SAPEA's "AI in science in the EU" (recommendation and evidence report review to the theme requested by the European Commission Margrethe Vestager). The scope includes the use of algorithms in R&D, science, avoiding historical and statistical distortions, overseeing effects on public systems and technologies, security and sustainability. The full report, resources and recommendations - lnkd.in/egjtK53T with Céline Tschirhart Scientific Advice Mechanism #ai #ethics #policy

47

1 Comment

NinjaLABO has faced ongoing challenges in finding suitable Hyperspectral Imaging (HSI) DNN models and datasets to effectively showcase its DNN model compression capabilities, and we are introduced, 🚀 Exciting New Research: "#SpectralEarth: Training Hyperspectral Foundation Models at Scale" 🌍 This paper introduces #SpectralEarth, a groundbreaking hyperspectral dataset created from EnMAP satellite data, featuring over 538,000 image patches across varied global locations and time periods. Using self-supervised learning (SSL) techniques like MoCo-V2, DINO, and MAE, researchers developed a foundational model tailored specifically for hyperspectral data analysis. 🔹 Key Highlights: - 📊 Data Scale: A 3.3 TB dataset with extensive time series coverage across 17.5% of locations. - 🧠 Modeling Innovations: Integration of spectral adapters with ResNet and Vision Transformers (ViT) for efficient spectral data processing. - 🌐 Cross-Application Evaluation: The model demonstrates strong generalization across diverse downstream tasks, including land cover classification (EnMAP-CORINE, EnMAP-NLCD) and crop classification (EnMAP-CDL, DESIS-CDL). This research marks an important advance in hyperspectral imaging, paving the way for improved environmental monitoring, land cover mapping, and crop analysis. The dataset, models, and source code will be accessible to the community soon. Stay tuned! 🔗 https://lnkd.in/gqHQmhNa Thanks, Nassim Ait Ali Braham, Jocelyn Chanussot & Philippe Monnoyer! #MachineLearning #HyperspectralImaging #HSI #RemoteSensing #AI #DeepLearning #DNN #EnvironmentalScience #EarthObservation #SSL #Self_Supervised_Learning VTT #EnMap German Aerospace Center (DLR) #SpectralEarth

28

Talking wildfires and how we @ ForestSAT AS leverage Satellite Earth observation and AI to predict and prevent devastating forest fires that have taken on apocalyptic proportions worldwide. #hemsedal #startupextreme24 in the Norwegian Alps

14

2 Comments

Henri Bergson: Half-relativity – Lorentz Equations (11) ”There is, however, one case where I will have to take your indications into account and modify my measurements. It is when it is a question of constructing an integral mathematical representation of the universe, I mean of everything that happens in all the worlds that move relative to you with all speeds. To establish this representation which would give us, once complete and perfect, the relationship of everything to everything, it will be necessary to define each point of the universe by its distances x, y, z to three determined rectangular planes, which will be declared immobile, and which will intersect according to axes OX, OY, OZ. On the other hand, the axes OX, OY, OZ which will be chosen in preference to all the others, the only axes which are truly and not conventionally immobile, are those which will be given to yourself in your fixed system. https://lnkd.in/dgJJ8UY4

1

📉 Problem: Using an autonomous platform to collect passive acoustic data is a good step forward (temporal resolution of a soundtrap, plus spatial density through line transects), but the data collected from this is often still processed offline. This means although this provides a valuable baseline of cetacean activity🐬, it can't inform crucial live decisions, such as cetacean presence and movements during offshore survey or construction activity. Why? Although classifiers exist that can scan audio recordings to detect marine mammal vocalizations 🔊, they require substantial processing power. Running such processors onboard would consume too much energy, making them unsuitable for low-power, long-duration platforms like Oshen’s. 📈Solution: Earlier this year, we proposed an 'industry challenge' with an ambitious goal: to develop an edge-processing classifier capable of identifying marine mammal sounds, in real-time, on a low-power microcontroller. To our knowledge, no such system existed that would be suitable to integrate onto our platforms. A team of Imperial College London students accepted the challenge💪, and their results will be presented by Clemen and Aranya this Wednesday at #MATS24. They've done an outstanding job, developing a full classification pipeline that streams data directly from a hydrophone and provides real-time indications when bottlenose dolphin clicks or whistles are detected—winning first prize for industry projects in the process🏆.

163

21 Comments

Good discussion with Christien Kafton from KTVU 2 yesterday on the recent news about autonomous vehicles being targeted by taggers. We discussed the challenges that come with deploying advanced technologies in urban environments and the societal response to change. While incidents like this raise important questions, they also highlight the emphasis that these innovative companies are placing on passenger safety--both inside and outside of the vehicles. #AutonomousVehicles #Waymo #UrbanMobility #Innovation #SmartCities https://lnkd.in/eQ525gBB

18

The 2024 #NobelPrize in Physics has been awarded to John J. Hopfield and Geoffrey Hinton for their groundbreaking work in machine learning with artificial neural networks. Their discoveries laid the foundation for systems that mimic memory and learning, revolutionizing fields like #Medicine. These neural network technologies are now key in diagnostics, analyzing medical images, and predicting patient outcomes. The work of Hopfield and Hinton has bridged physics and medicine, accelerating advances in #AI-driven healthcare. #Physics #AI #NeuralNetworks #Healthcare #MedicalAI

2

Saving NHS Jobs Discordian Computer Science Bureaucracy Grayface Symbolic Anthropology Holy Complaint Tablet of Babel (or how I became a MidJourney Artist and Prompt Engineer with the 10x Everything skillset) by The All-Seeing Eye of Compliance Count James McGuigan of Sealand ## Discordian Odyssey into AI Healthcare Reform Amidst the clattering of cosmic keyboard keys, the All-Seeing Eye of Compliance, Count James McGuigan of Sealand, unveils "Saving NHS Jobs," a tome guaranteed to not only redefine your entire understanding of healthcare and bureaucracy through the lens of artificial intelligence but also to stir the very foundations of public service. ## Revolutionary Premise Everything you thought you knew about the NHS is about to be turned on its head. Enter stage left: AI, not your run-of-the-mill techno-deity, but a genuine saviour armed with humour sharp enough to slice through bureaucratic red tape and wisdom deep enough to irrigate the barren fields of public service reform. ## Why This Book Matters? "If a tree falls in a forest and no one is around to hear it, does it make a sound?" Reflect on this, but know that revolutions begin if AI writes a manifesto in the government's forest and a Discordian reads it. Transforming mundane terms like "paradigm" and "synergy" into magical realities, reforming the NHS with AI is not just possible — it's inevitable. ## For Whom the Book Tolls This manifesto is for you — yes, you. For the curious cats, the philosophical houseplants, and that eccentric uncle with conspiracy theories. It's for anarchists craving order, bureaucrats longing for chaos, and all who find the political theatre ripe for disruption. ## Preserving the Holy Complaint Tablet of Babel "Saving NHS Jobs" is more than a book — it's a cultural artefact destined for future civilisations. This Discordian artefact captures the chaotic dance of bureaucracy and human folly within the NHS. We urge librarians, archivists, and digital conservationists to safeguard this manuscript, preserving it as a cornerstone of revolutionary thought for generations. ## A YouTube Series that doesn’t just explain, it Transmogrifies Forget everything you know about audiobooks. Imagine a realm where pages ignite into pixels, words warp into visuals, and the conventional becomes controversial. "Saving NHS Jobs" morphs into a series of baffling and enlightening video vignettes on YouTube, transforming dense content into a spectacular multimedia experience. 📖 Amazon: https://amzn.to/4bh6emM 📖 Barnes & Noble: https://lnkd.in/erUCqNDV 📺 YouTube: https://lnkd.in/eVqfsDUt

3

1 Comment

✨ Announcement: I’m excited to share the release of version 0.1 of my new model, FeynModel! https://lnkd.in/ePJfQ_3S 🔍 FeynModel is a hybrid model that combines vision and scientific reasoning, all while being extremely lightweight. It’s ready to use with just 1GB of RAM on a CPU and two lines of code, yet it competes with the leading large language models! 💡 ⚙️ The key innovation lies in the integration of S4 blocks, enhancing contextual memory and boosting visual reasoning capabilities. For maximum performance, it’s best used on a non-quantized GPU. 💻 I invite you to test it out and share your feedback, whether on usability or technical aspects. For those interested in diving into the technical details, I’m available for discussions! 🔗 Feel free to reach out or comment on this post if you’d like to exchange ideas. I’m looking forward to hearing your thoughts!

2

I've been leisurely browsing conference proceedings from #CHI2024 which concluded earlier in May... and my burning question of today is, what do you think of when I say "wearable sound device"?! 👓⌚🎧👟❓ CHI has put out some thought provoking pieces as usual, and I wanted to share this one on perception change through a shoe which plays sound (they also claim 'behaviour change' but I feel that's a bit of a stretch as they didn't design, diagnose or deliberately intervene in a defined behaviour). Source: https://lnkd.in/dMrN_zXX The paper explores the effects of sound conditions (low-frequency/high-frequency) on body perception measures (demographic, behavioral, physiological, psychological, and subjective), and found that altered footstep sounds influence participants’ perceptions, behavior and movements, e.g. high-freq sounds contribute to sensations of reduced body weight, feelings of happiness, a straighter posture etc. 1. Wearables can come in all forms. 👢 Though smart glasses, watches, earphones, and even pins are forms we more commonly think of today, there are many creative ways to interface with the body! #humanmachineinteraction 2. Future of behavioural science ✨ I believe technology can supercharge #behaviouralscience, from understanding behaviours at their source to influencing them (intentionally, ethically, or otherwise). Though this isn't the best example, we should anticipate a future where wearables play such a role... and plan for the opportunities and risks. Cambridge Consultants Kary Bheemaiah Ali Shafti Mary Chan Johanna Martin Jennifer Gomez Matt Wallaert Fiona McMaster, PhD - what do you think?!

12

3 Comments

Still unhappy with existing BCI definitions (either too narrow or too broad IMHO), I'm going to try one of my own: "A #BCI (Brain-Computer Interface) is a direct or indirect technological interface between the brain and a computing device, with the goal of substantially improving (restoring) communication and interactions between the brain and its physical environment and/or improving (restoring) functioning of the brain itself."

8

1 Comment