symisc_devel

There is also the PixLab API which let you automatically apply a blur filter for each detected face or any other target regions you want using only two Rest API endpoints.

Python: https://gist.github.com/symisc/6ecdea76ba0d33d73ea7f23cade0d...

PHP: https://gist.github.com/symisc/d54808915093e5375fdcb841e4365...

Docs: https://pixlab.io/cmdls

You are wrong. Symisc is an independent software vendor with a distributed team of ~14 collaborators and offices in Texas, Luxembourg and Tunis.

There is nothing to purchase at all if you want to train your own realnet models on the CPU such as Pedestrian, Car, etc. Just follow the documentation at https://sod.pixlab.io/api.html#sod_realnet_trainer on how to do so.

We do offer a custom version of our OCR reader app for Vision Impaired persons. The app require minimal interaction with the user. Just a single will trigger the phone camera, you have to take a picture and the text language is automatically recognized and played back aloud in your default language and accent with built-in translation powered by Google.

The OCR is powered by PixLab[1] and is damn good. The application homepage is https://i2s.symisc.net. If you are interested in trying the Vision Impaired version, let me know via chm at symisc dot net.

[1] https://pixlab.io/cmd?id=docscan

The SQLite source tree and DRH code in general are truly piece of art. Almost every line of code is carefully commented. Despite the complexity of the project, you'll learn a lot of practical concepts including expression tree generation, bytecode execution, how to test your code and so forth.

Did anybody here played with Palantir's software stack? If so, how is the overall experience especially the UI?

https://paddle.com do handle subscriptions very well with a good margin for startups.

The SOD embedded CV library do also implements a Real-time frontal face detection via its Realnets architecture (~5 ms on HD video stream).

https://sod.pixlab.io/c_api/sod_realnet_detect.html

https://github.com/symisc/sod/blob/master/samples/realnet_fa...

> Redis Labs CEO Ofer Bengal told me the company’s isn’t cash positive yet. He also noted that the company didn’t need to raise this round but that he decided to do so in order to accelerate growth. “In this competitive environment, you have to spend a lot and push hard on product development,” he said.

I'm always amazed how startups like this where R&D practically cost nothing are not profitable yet after years in the market and thousands of customers. C'mon guys 99% of the work is already done by antirez and the open source community.

Do you have any resources on how to allow ssh access from certain IP ranges? I can't find anything useful on this topic.

Installing a browser by default that cannot be uninstalled or even disabled then claiming it's the most installed browser, is a a little bit exaggerated. It's like Microsoft claiming that IE was the most installed browser on PCs, even if a great number of people only ever used it to download Firefox or Chrome.

https://art.pixlab.io does something similar using ML oriented approach (decision tree).

park.io is making $120K per month using an automated set of scripts that catch up pendingDelete domains from various registries.

Not that much on modern Linux/BSD systems at least. The kernel is smart enough to just copy fewer pages when forking not the entire parent process image.

SQLite already implemented such mechanism[0] that determines whether or not a given identifier is really an SQL keyword. The result is a single static hashtable (C array of chars) with O(1) lookup result. This is the fastest known implementation for keyword lookup.

[0]: https://sqlite.org/src/artifact/1f7f2ac1d9f262c0

Pure profit after taxes apparently.

The consensus we have reached at our company is as follows:

1. If you plan to write a software library [0] to be embedded directly onto foreign piece of code (i.e a host application), then go blindly with ANSI C or C99 at last resort. Not only writing bindings for your code should be straightforward but you will also benefit from the portability of C that make your code virtually available on every platform or architecture out there. Remember write once, run everywhere. Well, C fits exactly onto this category.

2. If on the other side you plan to write a standalone software such as a Desktop app, a network server, a game, etc. then modern C++ should be your first choice. You'll largely benefit from shared/unique pointers, built-in threading, lambdas, high performance data structures with relative portability accros modern systems. C++11/14/17 is a joy to program with once you master the basics.

[0]: https://sod.pixlab.io

Edit: formatting.

You may find SOD interesting for your embedded Computer Vision tasks. It is written in pure C and have an API similar to OpenCV.

https://sod.pixlab.io

Project homepage: https://github.com/symisc/ascii_art

Hey all. I worked on the UnQLite development team back into 2013 ~ 2014. It was designed to be the primary storage for a home targeting router manufactured by a French tech conglomerate.

The API is broken down into two independent layers plus one for various utility stuff like cross platform mmaped files, RC4 random numbers generation and so on. The first layer serve as a general purpose key/value store for the host application blobs (i.e binary or text data). You can store whole disk files inside a single database to generate a cross platform TAR archive for example.

The second one is the document store layer similar in concept to what MongoDB offer but without the networking layer since everything run in the same process of the host application. Both layers are transactional and are able to recover after some external failure thanks to the SQLite journaling mechanism from which UnQLite is based on.

Since its release in 2013, there was four major bugs including two critical that involves data loss under certain load. All known data corruption bugs were fixed in the last release and no major bugs were discovered a year or so.

The library enjoy particular success among Python and C/C++ developers especially Chinese that used to bombard the original developers with various internal aspect of the library.

Not that much. Contact them and they will bring back with an offer adequate to your dataset and budget.

You may take a look to SOD embedded project which is open source and do support facial detection at Real-time.

https://github.com/symisc/sod

PixLab do offer quota based monthly plans for such task including standard face detection, shape extraction, gender, age and emotion pattern extraction. They do charge $0.9 per 1000 requests after you reach your monthly quota (1.1M API calls).

https://pixlab.io/cmd?id=facedetect.

For those interested, PixLab let you analyze 50K images or video frames via its /nsfw endpoint for $25. They charge $0.9 per 1000 requests after you reach this quota.

https://pixlab.io/cmd?id=nsfw

https://pixlab.io/pricing

Sounds cool. Wat tech do you use for the streaming part?

Yes, they really don't care about any open source license. In fact, they are notorious for GPL violation.

What they were looking for is the Real-time face detector implemented directly in the C code which is available only for the commercial version of the library (https://pixlab.io/downloads).

We were recently approached by a contractor who works behalf the Chinese gov. Basically, they wanted to purchase an insane number of commercial licenses of our embedded computer vision library[1] to implement face tracking (detection and recognition) for a custom OpenBSD fork running on a homegrown CPU architecture with 512MB of RAM. They didn't tell more about this hardware but impression is that is the new generation of street Cameras.

[1]: https://github.com/symisc/sod

Technically speaking, face recognition has been a solved problem since the early 2000s despite facing few challenges nowadays including non-frontal, tiny or partial real-time detection. The most common algorithm is to output the bounding box (i.e. rectangle) coordinates for each present face, extract the facial landmarks (nose, pupils, mouth, etc.) points for the target face (the more landmarks you have, the more accurate the result is), perform some Euclidean distance calculation (sort of hash) and compare this to the existing face hashes (i.e. police database). The highest score is considered a potential match.

You can develop your own face recognition software providing a good dataset (i.e CelebA, LFW or the fed dataset if you have access to) using our embedded computer vision library[1] that we just released earlier this month or using the PixLab /facecompare HTTP endpoint (https://pixlab.io/cmd?id=facecompare).

[1]: https://github.com/symisc/sod

Yes, CV is always the first pass before character extraction. It envolve a lot of image processing routines including morphological operation like dilation, noise removal, etc. and finally blob detection[1]. All of the preprocessing routines are implemented in SOD.

[1]: https://sod.pixlab.io/c_api/sod_image_find_blobs.html

We released a C/C++ computer vision library (Python binding to follow) earlier this month which let you perform face recognition at Real-time from IoT or Raspberry PI devices. The library is dependency free, cross platform and should compile fine on most modern architectures with a C compiler.

https://github.com/symisc/sod