Discover the hidden secret in your iPhone's camera emoji. Using AI tools, we uncover an intriguing detail that Apple has subtly included since 2017 in this icon 📷 In this video, we: • Investigate the unusually detailed camera emoji • Utilize AI-powered image restoration techniques via s-i-m-u-l-a-t-i-o-n of the world • Explore Apple's emoji design process • Demonstrate how video generation models can reconstruct obscured data Our journey involves reaching out to original emoji designers and internet researchers, including: • Angela Guzman and Raymond Sepulveda, early iOS emoji creators https://medium.com/@agzmn/the-making-of-apples-emoji-how-designing-these-tiny-icons-changed-my-life-16317250a9ee We use various AI tools in our investigation: • https://neural.love for image upscaling • Gen-3 by Runway for video generation (runwayml.com) Join me on this tech-meets-detective journey as we use AI to uncover hidden secrets in the emojis we use every day. It's like CSI, but for your phone's keyboard! Subscribe for more content on AI applications and tech discoveries.

Generate AI portraits for free: https://neural.love/portraits You can enhance any video online here: https://neural.love/ Open source colab version of portrait generator tool made by the community: https://colab.research.google.com/drive/15j84iloYieB295OdAzs1ra1fSx65d00x (Make sure to copy it to your own Google Drive before starting your work) Sorry for a long time without new videos, potential lawsuits throw me off track for a while. ʕ·͡ᴥ·ʔ

🚀 I have boosted the original NASA video to 60 fps and upscaled it to 4K, here is the source NASA video: https://youtu.be/4czjS9h4Fpg "NASA's Mars 2020 Perseverance mission captured thrilling footage of its rover landing in Mars' Jezero Crater on Feb. 18, 2021" ⚠ Sound in this video is not real. For ambient sound I have used this amazing video of Curiosity Descent: https://youtu.be/Esj5juUzhpU #mars #Perseverance

A few months ago I found an amazing YouTube channel with some Lumiere films uploaded with great source quality. It was hard to resist not doing some Machine Learning magic with them, aaaand here is the result of a few months of processing and sound editing. Films used in this video: 🎞1895 France, Lyon, place des Cordeliers, https://catalogue-lumiere.com/place-des-cordeliers/ Factory outlet, France, Lyon, Monplaisir, chemin Saint-Victor (today rue du 1er Film) https://catalogue-lumiere.com/sortie-dusine-i/ The Landing of the photography congress in Lyon, https://catalogue-lumiere.com/debarquement-du-congres-de-photographie/ 🎞1896 Launch of a ship, France, https://catalogue-lumiere.com/lancement-dun-navire/ Switzerland, Geneva, National Exhibition, Swiss Village, https://catalogue-lumiere.com/rentree-a-letable/ Westminster Bridge, Great Britain, London, https://catalogue-lumiere.com/pont-de-westminster/ France, Lyon: Quai de l'Archevêché, https://catalogue-lumiere.com/lyon-quai-de-larcheveche/ Panorama of the Grand Canal taken from a boat, Italy, Venice, Grand Canal, https://catalogue-lumiere.com/panorama-grand-canal-venise/ Arrival of a train in Perrache station, France, Lyon, https://catalogue-lumiere.com/arrivee-dun-train-a-perrache/ Broadway, United States, New York, https://catalogue-lumiere.com/broadway/ 🎞1897 Jaffa Gate: east side, Jerusalem 1897 https://catalogue-lumiere.com/porte-de-jaffa-cote-est/ The pyramids, Egypt, Giza https://catalogue-lumiere.com/les-pyramides-vue-generale/ Panorama of the Golden Horn, Turkey, Istanbul https://catalogue-lumiere.com/panorama-de-la-corne-dor/ Camel caravan, Jerusalem, https://catalogue-lumiere.com/caravane-de-chameaux/ France, Lyon, place du Pont https://catalogue-lumiere.com/lyon-la-place-du-pont/ Japan, Kyoto, Honshu, https://catalogue-lumiere.com/lutteurs-japonais/ 🎞1899 Biarritz: the beach and the sea, France, Biarritz, Grande Plage, https://catalogue-lumiere.com/biarritz-la-plage-et-la-mer/ Bad weather at the port, Italy, https://catalogue-lumiere.com/mauvais-temps-au-port/ 🎞1900 View from a whaling boat in motion, France, Hyères https://catalogue-lumiere.com/vue-prise-dune-baleiniere-en-marche/ Panorama taken from a sedan chair, January 25, 1900, French Indochina (now Vietnam), village of Namo, Annam https://catalogue-lumiere.com/village-namo-panorama/ 🎞1902 Fort-de-France: market, Martinique, Fort-de-France, French Antilles https://catalogue-lumiere.com/fort-de-france-marche/ 🎶 Ending music used: Amosdoll Music channel https://youtu.be/b0c4f8alCCo

Source video: https://www.bbc.co.uk/programmes/p00fwcbn Recently I found an amazing interview with Yuri Gagarin in black and white from July 1961. It was done by the BBC during Gargarin’s 4-day visit to Great Britain as part of a Soviet Exhibition at Earl's Court in London. Gagarin became the first human in space on April 12, 1961 when he orbited the Earth for 108 minutes in the Vostok spaceship. Because Gargarin had a great smile and amazing charisma, and because I myself deeply love any space topics with space exploration, I couldn’t resist enhancing this video. 🚀 Denoised; 🚀 Colorized; 🚀4k upscaled; 🚀 60 fps boosted; 🚀 Facial restoration applied. Music used in this episode: https://freesound.org/people/frankum/sounds/442285/ https://freesound.org/people/PatrickLieberkind/sounds/396458/ #gagarin #space #neuralnetworks #upsacle #60fps

🔗 Source video from The Museum of Modern Art channel: https://youtu.be/2Ud1aZFE0fU 💖 Huge and sincere thank you to MoMA for making it available online in such great quality. 🎞 Upscaled with neural networks footage of "Wuppertal Schwebebahn" shot in 1902. Btw, the train system hasn‘t changed much and still functional. The "Schwebebahn" was built mostly over the river to save space. 💌 You can reach me here: https://neural.love/ ✔ Upscaled to 4K; ✔ FPS boosted to 60 frames per second, I have also fixed some playback speed issues; ✔ Stabilized; ✔ Colorized – please, be aware that colorization colors are not real and fake, colorization was made only for the ambiance and do not represent real historical data. 🎀 An interesting fact about that train system: Tuffi was a female circus elephant that became famous in West Germany during 1950 when she accidentally fell from the Wuppertal Schwebebahn into the River Wupper underneath. On 21 July 1950 the circus director Franz Althoff had Tuffi, four years old, to travel on the suspended monorail in Wuppertal, as a publicity stunt. The elephant trumpeted wildly and ran through the wagon, broke through a window and fell ~12 meters (39 ft) down into the River Wupper, suffering only minor injuries. A panic had broken out in the wagon and some passengers were injured. Althoff helped the elephant out of the water. Both the circus director and the official who had allowed the ride were fined. Source: https://en.wikipedia.org/wiki/Tuffi ______ 🎶 Song: Deluge by Cellophane Sam https://freemusicarchive.org/music/Cellophane_Sam/Sea_Change/01_Deluge #Wuppertal #Germany #Upscale #60fps #4k #machinelearning

Source video (with ambiance sound) – please subscribe to guy jones channel, he is doing an amazing job in ambiance sound adding: https://youtu.be/BYOGClmmeao 🎞 Upscaled with neural networks footage from the dawn of film taken in Tokyo, Japan from 1913-1915. 💌 You can reach me here: https://neural.love/ ✔ FPS boosted to 60 frames per second, I have also fixed some playback speed issues; ✔ Faces are enhanced too – I have added to the pipeline of algorithms a neural network which is specially designed for facial restoration. ✔ Image resolution boosted up to 4k – with digital artifacts, but some parts are improved noticeably; ✔ Removed noise and fixed some damaged parts. ✔ Colorized – please, be aware that colorization colors are not real and fake, colorization was made only for the ambiance and do not represent real historical data. #Tokyo #Japan #Upscale #60fps

In this episode of A.I. stuff - I’ve applied a bunch of different neural networks in an attempt to generate realistic faces of people from famous paintings. ...

In honor of the end of World War II in Europe I have attempted to stabilize and deblur an amazing archival video from @BERLIN CHANNEL: https://youtu.be/R5i9k7s9X_A (it is already with sound) That’s how Berlin looked just after the German surrender. Video shot in color shows the situation in the city in July 1945 and residents’ daily life among the ruins. Unfortunately, the source quality is quite poor, which is why you can see some glitches and shakiness in the final result. While this example is not as strong as others on this channel, I have included my efforts here to show what our current limitations are and how much remains to be discovered in the field of ML—our progress towards an ideal goal is sometimes a long and arduous process, but we should not give up. I have applied the following algorithms: – Deblur – Deflickering – Stabilization – FPS boosting – Speed correction – Upscale _____ #Berlin #WW2 #Upscale #60fps

Upscaled with an ensemble of neural networks Louis Le Prince's footage, Roundhay Garden Scene, shot in October 1888. I have used this source for upscaling: https://collection.sciencemuseumgroup.org.uk/objects/co8646638/glass-copy-negative-of-roundhay-garden-scene-glass-negative 💌 You can reach me here: https://neural.love/ ______ #upscale #Leeds #1888 #old #England

Upscaled with neural networks trip down Market Street, San Francisco, 1906. This film was shot on April 14, 1906, just four days before the San Francisco earthquake and fire which killed an estimated 3 000 people and destroyed over 80% of the city of San Francisco. It was produced by: Harry, Herbert, Earle and Joe. Harry J. 💌 You can reach me here: https://neural.love/ ✔ FPS boosted to 60 frames per second (DAIN); ✔ Image resolution boosted up to 4k (ESRGAN); ✔ Improved video sharpness; ✔ Noise removal; ✔ Colorized* *Please, be aware that colorization colors are not real and fake, colorization was made only for the ambiance and does not represent real historical data. Source video (with ambiance sound): https://youtu.be/v2u4J7Dbouo ______ Please, keep in mind that 4k resolution playback is mostly not available on the phones. #upscale #SanFrancisco #1906 #old #usa

Upscaled with neural networks footage of Amsterdam in the Netherlands in 1922. A lot of crowd scene and some people dancing ✨ 💌 You can reach me here: https://neural.love/ ✔ FPS boosted to 60 frames per second; ✔ Image resolution boosted up to 4k; ✔ Improved video sharpness; ✔ Stabilization; ✔ Colorized* *Please, be aware that colorization colors are not real and fake, colorization was made only for the ambiance and do not represent real historical data. Source video (with ambiance sound) – please subscribe to Guy Jones channel, he is doing an amazing job in ambiance sound adding: https://youtu.be/Kh4eVoqUoXo ______ Please, keep in mind that 4k resolution playback mostly not available on the phones.

Upscaled with neural networks footage from the dawn of film taken by Mitchell and Kenyon in North England, 1901. In the video various films taken in 1900-1901 displaying some of the grittier nature of work in those days. As you can see, source quality is really important to make a descent upscale video. 💌 You can reach me here: https://neural.love/ ✔ FPS boosted to 60 frames per second; ✔ Image resolution boosted up to 4k – with digital artifacts; ✔ Improved video sharpness; ✔ Colorized: Please, be aware that colorization colors are not real and fake, colorization was made only for the ambiance and do not represent real historical data. Source video (with ambiance sound) – please subscribe to Guy Jones channel, he is doing an amazing job in ambiance sound adding: https://youtu.be/jfN60ZlZhpY List of films in this video: Miners Leaving Pendlebury Colliery / https://player.bfi.org.uk/free/film/watch-miners-leaving-pendlebury-colliery-1901-1901-online Workforce Leaving Alfred Butterworth and Sons, Glebe Mills, Hollinwood (1901) / https://player.bfi.org.uk/free/film/watch-workforce-leaving-alfred-butterworth-and-sons-glebe-mills-hollinwood-1901-1901 Workforce of Parkgate Iron and Steel Co., Rotherham (1901) / https://player.bfi.org.uk/free/film/watch-workforce-of-parkgate-iron-and-steel-co-rotherham-1901-1901-online ______ Please, keep in mind that 4k resolution playback mostly not available on the phones.

Upscaled with neural networks footage from the dawn of film taken in Belle Époque-era Paris, France from 1896-1900. 💌 You can reach me here: https://neural.love/ ✔ FPS boosted to 60 frames per second; ✔ Image resolution boosted up to 4k – with digital artifacts, but some parts are improved noticeable, I'm preparing new dataset for that process, 4k is till in beta; ✔ Improved video sharpness; ✔ Colorized – I'm still unsure about it, but regarding to high request from the subscribers I decided to use DeOldify NN again on this video. If you don't like how DeOldify doing colorization, please let me know in the comments, I will upload b&w version and put a link here. ⚠ Please, be aware that colorization colors are not real and fake, colorization was made only for the ambiance and do not represent real historical data. Source video (with ambiance sound) – please subscribe to guy jones channel, he is doing an amazing job in ambiance sound adding: https://youtu.be/NjDclfAFRB4 ______ Please, keep in mind that 4k resolution playback mostly not available on the phones.

Upscaled with neural networks 1896 Moscow footage taken in May by Charles Moisson from the Lumière company. 💌 You can reach me here: https://neural.love/ Ensemble of neural networks used for: 1) FPS boosting – to 60 frames per second; 2) Image resolution boosted up a bit with ESRGAN (general dataset); 3) Resorted video sharpness, removed blur, removed compression "artefacts"; 4) Colorized (optional) – due to high request I have decided to include both versions of the processed video: colorized and black and white. ⚠ Please, be aware that colorization colors are not real and fake, colorization made only for the ambiance and do not represent real historical data. Source video (with ambiance sound): https://youtu.be/FYHF3a9R5DY ______ Восстановленная версия видео от 1896 года снятого Чарльзом Муассоном, компания Люмьер. Ансамбль нейронных сетей использован для: 1) Увелечения кадров в секунду до 60; 2) Немного улучшил качество картинки увеличив разрешение каждого кадра, не сильно помогло, но мне понравилось; 3) Другая нейронка позволила удалить шум, размытие, артефакты исходного видео; 4) Еще одна нейронка покрасила видео, мне не очень нравится как она красит, но люди просят – поэтому я включил улучшенную ЧБ версию в это же видео, помимо покрашенного. Оригинал видео взят тут (Озвучку я тоже взял у этого автора, я ее не сам сделал): https://youtu.be/FYHF3a9R5DY

Upscaled with neural networks 1911 New York footage taken by the Swedish company Svenska Biografteatern on a trip to America: 💌 You can reach me here: https://neural.love/ ✔ FPS boosted to 60 frames per second (DAIN); ✔ Image resolution boosted up to 4k (ESRGAN, custom weight); ✔ Resorted video sharpness (AE and other plugins); ✔ Colorized – I'am still unsure about this, but regarding to high request from the subscribers decided to test DeOldify NN on this video. ⚠ Please, be aware that colorization colors are not real and fake, colorization made only for the ambiance and do not represent real historical data. Source video (with ambiance sound): https://youtu.be/aohXOpKtns0 ______ Please, keep in mind that 4k resolution playback mostly not available on the phones.

Upscaled and resounded version of Apollo 16 mission "Grand Prix" part: FPS boosted to 60 frames per second; Image resolution boosted up to 4k. 💌 You can reach me here: https://neural.love/ Source for upscaling: https://youtu.be/X30z82aeSHw ______ Please, keep in mind that 4k resolution playback mostly not available on the phones.

Generated with StyleGAN2 music video. ___________ Нейронка генерирует лица в так битам, а рот "подпевает" под вокал. Используется StyleGAN2 + https://github.com/deezer/spleeter Если вы программист, то самому поиграться можно тут: https://colab.research.google.com/github/Norod/my-colab-experiments/blob/master/WikiArtCultureShock.ipynb ______ За идею к видео спасибо: http://rolux.org

Upscaled and resounded version of a classic B&W movie: Arrival of a Train at La Ciotat, The Lumière Brothers, 1896 💌 You can reach me here: https://neural.love/ Source used to upscale: https://youtu.be/MT-70ni4Ddo Algorithms that were used: ››› To upscale to 4k – Gigapixel AI – Topaz Labs https://vc.ru/76580 ››› To add FPS – Dain, https://sites.google.com/view/wenbobao/dain ⇒ Update Colorized by DeOldify Neural Network version of this video: https://youtu.be/EqbOhqXHL7E ⇒ Update 2 Articles about project: https://gizmodo.com/neural-networks-upscale-film-from-1896-to-4k-make-it-l-1841449875 https://www.engadget.com/amp/2020/02/04/how-ai-helped-upscale-an-antique-1896-film-to-4k/ https://boingboing.net/2020/02/04/1896s-arrival-of-a-train-a.html https://www.theverge.com/tldr/2020/2/5/21124838/classic-1896-short-film-upscaled-4k-neural-networks-arrival-la-ciotat-station #1896 #upscale #Train

The artificial intelligence of AIVA followed the motifs from Dvořák's works and completed one of his musical sketches. This brought a three-movement composition called in honour of the master "From the Future World. Recorded by me. ------ Есть такой композитор – Антонин Дворжак и он умер 115 лет назад не дописав одно из своих произведений «Из мира будущего» (From the Future World). Но спасибо алгоритмам и стартапу AIVA, произведение дописали и оркестр Prague Philharmonia в ноябре исполнили заключительную часть пьесы.

Playing with http://plink.in/ and bird is listening.

A popular photo of Buzz Aldrin from the moon, where he stands on the moon, and in the reflection of his helmet you can see Neil Armstrong who photographs him 💫

Поигрался с демо-версией примерки одежды в "дополненной реальности". Скачать можно тут: https://testflight.apple.com/join/TOPQxoP8

Army of Lovers - When The Night Is Cold. Me – at the piano 💖