Viral Hooks
Trending short videosTesting tech hardware to its limits
D

@dataslayer

TikTok · Talking Head Edit

@dataslayer Coral.ai Edge TPU Supercharges Your Raspberry Pi! #techtok #raspberrypi #raspberrypi4 #coralai #tpu #machinelearning #tpu #tensorflow #coding #python #algortihm #artificialintelligence #Tech #computervision #ai #opencv #vscode

View original on TikTok

The edit1 shot · 23 text
Shots Text0:59

Performance

Views
150.6K
Likes
9.3K
Comments
59
Engagement
6.2%
Length · 26 Mar 2023
0:59

Video #5321 · Talking Head Edit

Testing tech hardware to its limitsfrom @dataslayer

[01]

Overview

The creator tests the limits of a Raspberry Pi by attempting real-time video object recognition. He introduces the Google Coral AI Edge TPU, shows the unboxing and setup process, runs a successful initial image classification test, and finally demonstrates the system successfully identifying an electric guitar in a live video feed.

[02]

Simple outline

  1. 1State an impossible challenge
  2. 2Introduce a 'magic' solution
  3. 3Show a preliminary success
  4. 4Deliver the final, decisive proof

[03]

Related videos

[04]

Classification

Concepts
ChallengeProduct DemoUnboxingQuest Recap
Elements
Text OverlayB-Roll CutawayScreen RecordingGraphic Overlay
Topics

[05]

Hook analysis

What stops the scroll in the first 3 seconds

Title hook · 15%

Today

Verbal hook · 75%

Today, I'm gonna push my Raspberry Pi to the absolute limit and find out if it can handle one of the most challenging computations in all of computer science.

Visual hook · 10%

The creator holds up a Raspberry Pi connected to a Google Coral AI Edge TPU, showcasing the specific tech hardware being tested against a high-rise city view background.

[06]

Hook template

Steal this for your own video

Today, I'm going to push [tool or device] to its absolute limit and find out if it can handle [impressive or difficult challenge].

  • — Today, I'm going to push a $35 microcontroller to its absolute limit and find out if it can handle real-time AI object recognition.
  • — Today, I'm going to push a free photo editing app to its absolute limit and find out if it can handle a full professional magazine cover shoot.
  • — Today, I'm going to push a budget home espresso machine to its absolute limit and find out if it can handle a world-class barista competition recipe.

[07]

Hook and payoff

The promise the opener makes, and how the video pays it back

Hook

State an extreme, high-stakes challenge that seems impossible for a familiar object or tool, immediately establishing a clear goal and a high degree of difficulty.

Payoff

Execute the ultimate challenge and deliver a clear, quantifiable, and satisfying success, resolving the initial hook with a definitive verdict.

[08]

Narrative framework

The structural pattern this video runs on

Framework

The Impossible Challenge

Frame a seemingly impossible task for a common tool, introduce a specialized 'supercharger' product as the solution, and then prove its effectiveness through a series of escalating tests.

  1. 1

    The Catalyst Introduction

    Introduce a new tool, product, or technique as the key to overcoming the previously established impossibility, framing it as a game-changer.

  2. 2

    The Power-Up Explanation

    Detail the capabilities of the new tool, often using hyperbole to frame it as a 'supercharger' that fundamentally alters the base item's potential.

  3. 3

    The Assembly Montage

    Show the procedural steps of implementing the solution, building anticipation by showing the 'work' being done before the test.

  4. 4

    The Preliminary Proof

    Conduct a smaller, simpler test that results in a quick, impressive success, validating the setup and building momentum towards the main event.

  5. 5

    The Final Question

    Explicitly restate the ultimate challenge, creating a moment of peak suspense and clearly defining the stakes for the final test.

Themes

  • Technology
  • Innovation
  • Problem-solving

Techniques

  • Product demonstration
  • Step-by-step process
  • Challenge and resolution

Trends

  • AI at the edge
  • DIY tech projects

Post caption

Coral.ai Edge TPU Supercharges Your Raspberry Pi! #techtok #raspberrypi #raspberrypi4 #coralai #tpu #machinelearning #tpu #tensorflow #coding #python #algortihm #artificialintelligence #Tech #computervision #ai #opencv #vscode

Text overlays

  • contextual · beginning

    Today

  • contextual · beginning

    Raspberry

  • contextual · beginning

    limit

  • branding · beginning

    Coral

  • contextual · beginning

    find out

  • contextual · beginning

    handle

  • contextual · beginning

    challenging

  • contextual · beginning

    in all

  • contextual · beginning

    science

  • branding · beginning

    humanoid

  • contextual · beginning

    unleashed

  • contextual · beginning

    everything

  • contextual · middle

    called

  • contextual · middle

    AI

  • contextual · middle

    TPU

  • contextual · middle

    supercomputer

  • contextual · middle

    i'm scared

  • branding · middle

    Coral.ai

  • contextual · end

    fast

  • title · end

    Real-Time Video???

  • contextual · end

    BOOM

  • informational · end

    91%

  • informational · end

    Inference: 6.84ms FPS: 0.5 91% electric guitar

Transcript

Today, I'm gonna push my Raspberry Pi to the absolute limit and find out if it can handle one of the most challenging computations in all of computer science. Video best recognition. No, normally this would be completely out of the question, but Google just unleashed a new product that's about to change everything. It's called a Coral AI Edge TPU and it's a piece of hardware designed to do only two things, look cool and run machine learning algorithms, which can take a relatively low powered device like a Raspberry Pi and turn it into a supercomputer. That escalated quickly. Worst, I'm gonna install the camera onto my Raspberry Pi, plug in the Coral AI, and boot it up. Then I'm gonna download some packages from coral.ai. I totally understand this code. Tier, we're gonna test out one of the pre built models and classify this picture of a bird. That was crazy best, but will it be able to handle real time video? Let's find out. So I found the video detection script and ran that puppy like I was Usain Bolt. Then I walked into the video with my guitar rank boom, the Raspberry Pi classified the object with an astounding 91% certainty.