Sunday, February 26, 2023

Lepton image compression: saving 22% losslessly from images at 15MB/s

This open-source project is no longer maintained or supported by Dropbox. Please refer to Lepton’s GitHub page for more information.

~ ~ ~

We are pleased to announce the open source release of Lepton, our new streaming image compression format, under the Apache license.

Lepton achieves a 22% savings reduction for existing JPEG images, by predicting coefficients in JPEG blocks and feeding those predictions as context into an arithmetic coder. Lepton preserves the original file bit-for-bit perfectly. It compresses JPEG files at a rate of 5 megabytes per second and decodes them back to the original bits at 15 megabytes per second, securely, deterministically, and in under 24 megabytes of memory.

We have used Lepton to encode 16 billion images saved to Dropbox, and are rapidly recoding our older images. Lepton has already saved Dropbox multiple petabytes of space.

Community participation and improvement to this new compression algorithm is welcome and encouraged!

Lepton at scale

At Dropbox, the security and durability of your data are our highest priorities. As an added security layer, Lepton runs within seccomp to disable all system calls except read and write of already-open file descriptors. Lepton has gone through a rigorous automated testing process demonstrating determinism on over 4 billion photos and counting. This means that once we verify an image decodes back to its original bits the first time, we can always get back to the original file in future decodes.

All of our compression algorithms, including Lepton, decode every compressed file at least once and compare the result to the input, bit-for-bit, before persisting that file. Compressed files are placed into kernel-protected, read-only, memory before the bit-for-bit comparison to guarantee they are immutable during the full verification process.

PackJPG v2.5k (01/22/2016)

packJPG is a compression program specially designed for further
compression of JPEG images without causing any further loss. Typically
it reduces the file size of a JPEG file by 20%.

Friday, February 24, 2023

Rare: Star Trek: Leonard Nimoy explains How Television Works 1977 vintage tech electronics CRT Spock

Remembering Leonard Nimoy - The following is an excerpt from a rare 1977 documentary on How Television Works, featuring Star Trek's Leonard Nimoy as the narrator and presenter.  The film is in full color with animation and excellent vintage footage of early television technology, including early video tape recording (VTR) machines.  The original film is about 22 minutes long.  We have obtained a 16mm print and will preserve it.  This is a 7 minute excerpt highlighting Leonard Nimoy’s narration of the technology behind early television up to the 1970’s.

Leonard Nimoy is known worldwide for playing Mr. Spock in the Star Trek franchise for nearly 50 years, beginning with the original Star Trek pilots and then 79 episodes from 1966 to 1969, through the Animated Series, numerous Star Trek films, and appearances in Star Trek: The Next Generation.   He also appeared in many TV series including The Outer Limits,The Man from Uncle, and in the cast of Mission Impossible (for 49 episodes).   
Among other things, he directed films Star Trek III: The Search for Spock, and the highly successful Star Trek IV: The Voyage Home. 
He was born on March 26, 1931 in the West End of Boston,  Massachusetts, and died on February 27, 2015, in California.

Wednesday, February 15, 2023

Draco - 3D Data compression..


Draco is a library for compressing and decompressing 3D geometric meshes and point clouds. It is intended to improve the storage and transmission of 3D graphics.

Draco was designed and built for compression efficiency and speed. The code supports compressing points, connectivity information, texture coordinates, color information, normals, and any other generic attributes associated with geometry. With Draco, applications using 3D graphics can be significantly smaller without compromising visual fidelity. For users, this means apps can now be downloaded faster, 3D graphics in the browser can load quicker, and VR and AR scenes can now be transmitted with a fraction of the bandwidth and rendered quickly.

Draco is released as C++ source code that can be used to compress 3D graphics as well as C++ and Javascript decoders for the encoded data.

Accidently stumbled on this when exploring 


How an iPhone Film Shocked the Movie World

This is the story about how an indie filmmaker used an iPhone, an anamorphic lens and FiLMiC Pro to get a major break into the industry.

Wednesday, February 08, 2023

Streamroot’s hybrid webRTC-based P2P video CDN

Streamroot: Powering robust, scalable video delivery for the World Cup
Streamroot uses Google Kubernetes Engine and Google Cloud Load Balancing to handle peaks in traffic for its global peer-accelerated content delivery solution that powers over 20 million video sessions a day with zero downtime.

Hybrid Peer-to-Peer Delivery

A visualization of Streamroot’s hybrid webRTC-based P2P video CDN integrated with JW Player in HTML5.  wrote a number of webRTC libraries, and also a major developer for streamroot, he's now with netflix.  

Tuesday, February 07, 2023

Synthesia is an AI video creation platform.

Is free? You can make a free demo video to test the AI generator (limited to 200 words). Simply upload or type in your script, and get your video.If you like the results, you can upgrade to the Personal pricing plan ($30/month) or a custom Corporate plan.


If you are running video around your home theater, you probably use HDMI. If you are running it in a professional studio, however, you are probably using SDI, Serial Digital Interface. [Chris Brown] looks at SDI and shows a cheap SDI signal generator for an Arduino.

On the face of it, SDI isn’t that hard. In fact, [Chris] calls it “dead simple.” The problem is the bit rate which can be as high as 1.485 Gbps for the HD-SDI standard. Even for a super fast processor, this is a bit much, so [Chris] turned to the Arduino MKR Vidor 4000. Why? Because it has an FPGA onboard. Alas, the FPGA can’t do more than about 200 MHz, but that’s fast enough to drive an external Semtech GS296t2 serializer which is made to drive SDI signals.

The resulting project contains the Arduino, the serializer, a custom PCB, and both FPGA and microcontroller code. While the total cost of the project was a little under $200, that’s still better than the $350 to $2000 for a commercial SDI signal generator.

If you want to play along, the files are out on GitHub. We used the Vidor back in 2018 when it first came out. If you need a quick start on FPGAs, there’s always our boot camp.