perception

A two-minute BBC Horizon clip demonstrating the McGurk effect, where what you see of a speaker's mouth changes the sound you hear.

The original invisible gorilla awareness test by Daniel Simons and Christopher Chabris, in which viewers count basketball passes and often miss a gorilla crossing the scene. It demonstrates inattentional blindness.

A BBC Horizon clip demonstrating the McGurk effect, where the same audio sounds like different syllables depending on the mouth movements you see. It shows that people integrate visual speech information into what they hear.

A player piano is made to reproduce human speech, here reciting a European Environmental Criminal Court proclamation in German at the 2009 World Venice Forum. This upload adds subtitles so the words become understandable.

Destin Sandlin of Smarter Every Day undergoes hypoxia training and experiences his own impaired thinking as oxygen is withdrawn. The video shows how quickly and unnoticeably oxygen deprivation degrades judgment, illustrating why you put your own mask on first.

The SIGGRAPH 2005 demo of dual photography from Stanford, which swaps the roles of projector and camera to render a scene from the light's point of view. It includes the reveal of a hidden playing card.

Andrew, diagnosed with early onset Parkinson's at 35, shows how deep brain stimulation surgery has controlled his motor symptoms. The short video shows his tremors with the stimulator off and on.

Destin of Smarter Every Day learns to ride a bicycle whose steering is reversed, taking about eight months, and compares it with how his young son learned much faster. It illustrates knowledge versus understanding and brain plasticity in adults versus children.

Mark Rober and Jabril build an app that uses machine learning to decode baseball sign sequences. The video also explains neural networks in a simple, accessible way.

The Thought Emporium grows living rat neurons on an electrode array and tries to train them to play Doom. It is part of a series on biological computing.

The Thought Emporium builds a homemade multi-electrode array, grows human neurons on it and attempts to take readings. It is the first part of a series aimed at connecting living neurons to a computer.

A ShortCircuit (Linus Tech Tips) review of Nvidia's RTX Voice, which uses GPU-based neural networks to remove background noise from microphone audio. It demonstrates how well the filtering works.

A University of Tokyo lab demonstrates a book scanner that flips pages automatically and scans over 250 pages per minute. It uses real-time 3D recognition of the flipped pages and then restores each one to a flat image.

Neural-network upscaled, 60 fps, stabilized and colorized version of 1902 footage of the Wuppertal Schwebebahn suspended railway in Germany. The source film comes from MoMA. It gives a vivid view of an early-1900s city and a train system that still runs.

SoundHound's CEO demos Hound, a voice assistant, handling fast compound natural language queries with context and detailed criteria. The video leaked before the product's official launch.

A SIGGRAPH Asia 2013 research demo of 3-Sweep, a tool for extracting editable 3D objects from a single photo using a few user strokes. It predates modern AI image tools.

Bobby McFerrin uses audience participation at a World Science Festival event to demonstrate how instinctively people follow the pentatonic scale.

Adobe's VoCo demo from the MAX 2016 Sneak Peeks, co-hosted by Jordan Peele, showing software that changes words in a voiceover by typing new ones.

Mark Rober tests whether self-driving cars can be fooled, comparing a lidar-equipped car with a camera-only Tesla in scenarios including a painted Wile E. Coyote wall and smoke.

A demonstration of Larynx, an offline open source neural text-to-speech system supporting nine languages. The video itself is narrated with a Larynx-generated voice.

Footage of FAA altitude chamber hypoxia training for pilots in which a participant passes out from oxygen deprivation. It shows how quickly and subtly cognition fails at altitude.

Footage shot in New York City in 1911 by a Swedish film company, upscaled to 4K and 60 fps with neural networks and optionally colorized. The creator notes the colors are artificial and added only for ambiance.

UploadVR covers a Facebook Reality Labs prototype that tracks facial expressions using cameras mounted in a VR headset. The updated version captures subtle micro-expressions for realistic avatars.

A short demo from the Ishikawa Komuro Lab of a high-speed robot hand paired with a 1000fps vision system performing dexterous manipulation. It shows what very low-latency vision and actuation can do.

A demo of PREDATOR (TLD), a real-time object tracker that learns the appearance of its target and improves while it tracks. The technology was later developed commercially.

Marques Brownlee records an uninterrupted drive using Tesla's Full Self Driving beta (v10.69) on ordinary roads. He shows how it handles everyday traffic and where he needs to intervene.

A short demo of the Deep Visualization Toolbox, which shows the activations inside a convolutional neural network as it processes live camera input. Viewers see individual units responding to edges, faces and text.

Yandex shows its self-driving taxi navigating snowy Moscow streets after a heavy snowfall, dealing with traffic, pedestrians, parked cars and snowbanks on real roads.

MIT researchers show that sound can be recovered from silent video by analyzing tiny vibrations in objects such as a potato chip bag. It works even with consumer cameras by using rolling shutter.

Drive.ai shows its car driving fully autonomously on a rainy night in Mountain View, with heavy downpour, glare, a four-way stop and a broken traffic light.

Microsoft Research's Rick Rashid demonstrates a speech recognition and translation system, built on deep neural networks, that turns his spoken English into Chinese speech in 2012.

Ashley Feniello demonstrates VimSpeak, a voice-controlled interface for Vim inspired by Tavis Rudd's voice programming talk. The demo includes working through a VimGolf challenge by speech.

AI Addict films a Tesla FSD Beta 8.2 drive through Oakland, annotating close calls, disengagements, pedestrians and wrong-lane moments with timestamps.

Jack Gallant's lab demonstrates reconstructing movie trailer clips a subject watched from fMRI brain activity, using regression models trained on hours of viewing data. The video shows the original clip beside the reconstruction.

A short video accompanying the CVPR 2018 paper Learning to See in the Dark. It shows a neural network producing sharp images from very low-light raw sensor data, compared against denoising approaches.

JJRicks rides in Waymo's fully driverless Level 4 cars in Phoenix with nobody in the front seats, across two long rides. It is the first in a series of such ride videos.

A short Tesla promotional video showing a car driving itself along a route from the company's Palo Alto headquarters, marketed as Full Self-Driving.

A University of Tokyo demo of a rock-paper-scissors robot that always wins, using high-speed vision to read the human hand and respond faster than people can perceive.

A short research demo of Autonomous Intersection Management, a protocol in which self-driving cars reserve slots through an intersection instead of obeying signals or stop signs. It uses UT Austin's autonomous vehicle Marvin and a simulation to show cars crossing without stopping.

A one-minute clip by cpddan reproducing how a Tesla on Autopilot 2.5 may have steered into a gore point barrier, as in a fatal Highway 101 crash.