Synergistic Face Detector

Closed weights NEC Laboratories,Courant Institute of Mathematical Sciences 16.6K parameters December 2004

No estimate

No hardware requirements for this model

The weights for this model have not been published, so it cannot be downloaded or run on your own hardware at any size. It is reachable only through its provider, and no graphics card changes that.

On record

Full specification

Everything on record for this model. Most of it describes how it was trained rather than how it runs — useful context for judging how much work went into it, and how it compares with models built at a different scale.

Origin

Who built this model, where, and when it was published.

Organisation
NEC Laboratories,Courant Institute of Mathematical Sciences
Organisation type
Industry,Academia
Country
United States of America
Published
1 December 2004
Authors
Margarita Osadchy, Matthew Miller, Yann LeCun

What it does

The problem areas the model was built for. A model can carry several of each.

Domain
Vision
Task
Face detection

Size

How large the model is and how much data it was trained on. Parameters are the figure that decides whether it fits on a given graphics card.

Parameters
16.6K

"We employ a network architecture similar to LeNet5 [6]. The difference is in the number of maps. In our architecture we have 8 feature maps in the bottom convolutional and subsampling layers and 20 maps in the next two layers. The last layer has 9 outputs to encode two pose parameters" Using the general architecture of LeNet5 from "Gradient-based learning applied to document recognition" (LeCun, 1998), we get: C1: CNN, D = 8, K=5, C=1; Parameters = 200 C3: CNN, D=20, K=5, C=8; Parameters = 4,00…

Training data
tokens

52,850x2x5

Epochs
9

Training compute

The arithmetic performed to train the model, measured in floating-point operations. It is a measure of what the training run cost, not of how fast the finished model answers you.

Training compute
5.3 × 10¹³ FLOP

26*60*60*[(2.81E+03)*((1E6)/2.5)]*0.5 "Training was performed using LUSH [1], and the total training time was about 26 hours on a 2Ghz Pentium 4" "The Progress of Computing" (Nordhaus, 2001) says that a Pentium 4 runs at 2.81E+03 MSOPS, and that "Most benchmarks find that 1 million flops correspond to between 2 and 3 MSOPS" A utilization rate of 0.5 was estimated since this is a smaller model that ran on a CPU (both of which suggest a slightly higher utilization rate than 0.3 to 0.4)

How it was established
Hardware

The training run

What it physically took to train: which chips, how many, for how long, and what that drew from the wall.

Training hardware
Intel Pentium 4 2.40
Wall-clock time
26 hours

"Training was performed using LUSH [1], and the total training time was about 26 hours on a 2Ghz Pentium 4"

Availability

Whether you can obtain the model and run it on your own hardware, which is what decides if any of the graphics-card figures on this page apply.

Weights
Closed — provider access only
Model access
Unreleased

How it is classified

Labels the source dataset applies when tracking notable models, and how confident it is in the entry.

Record confidence
Confident

Sources

Where this record came from and when it was last checked.

Reference
Synergistic Face Detection and Pose Estimation with Energy-Based Models
Last updated
28 November 2025

What the numbers mean

About this model

Synergistic Face Detector was published by NEC Laboratories,Courant Institute of Mathematical Sciences, in the country recorded as United States of America, during December 2004. The category the publisher falls under is industry,Academia.

It works in the domain of Vision, and is recorded as performing the task of face detection.

This is a closed model: the trained values stayed with whoever produced them, and there is no local version to run.

How it was trained

Producing it required arithmetic totalling around 5.3 × 10¹³ FLOP, on hardware recorded as Intel Pentium 4 2.40. That figure measures what producing the model cost, and has no bearing on how fast it answers.

Answers

Synergistic Face Detector — common questions

01

Synergistic Face Detector— is it open source?

No. Its weights have not been published, so it exists only as a service controlled by its owner.

02

Synergistic Face Detector— how many parameters does it have?

It has a parameter count of 16.6K. "We employ a network architecture similar to LeNet5 [6]. The difference is in the number of maps. In our architecture we have 8 feature maps in the bottom convolutional and subsampling layers and 20 maps in the next two layers. The last layer has 9 outputs to encode two pose parameters" Using the general architecture of LeNet5 from "Gradient-based learning applied to document recognition" (LeCun, 1998), we get: C1: CNN, D = 8, K=5, C=1; Parameters = 200 C3: CNN, D=20, K=5, C=8; Parameters = 4,000 C5: NN, D=20, K=5, C=20; Parameters = 10,000 F6: NN, N=20, M=84; Parameters = 1,680 Output: NN, N=84, M=9; Parameters = 756 Pooling layers: negligible Parameters = 16,636. That figure is the total, and it is what decides how much memory the model needs — roughly half a gigabyte per billion at the compression most people use.

03

Synergistic Face Detector— who created it?

It was published by NEC Laboratories,Courant Institute of Mathematical Sciences, based in United States of America, an organisation categorised as industry,Academia.

04

Synergistic Face Detector— when was it released?

It was published in December 2004. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

05

Synergistic Face Detector— what is it used for?

It works in the domain of Vision, and is recorded as handling the task of face detection. A model can carry several of each, so these are the areas it was built for rather than a limit on what it will attempt.

06

Synergistic Face Detector— how much compute was used to train it?

Training consumed around 5.3 × 10¹³ FLOP, on hardware recorded as Intel Pentium 4 2.40. That measures what producing the model cost and says nothing about how quickly it answers once trained — inference speed comes from memory bandwidth, not from the training budget.

07

Synergistic Face Detector— what GPU do I need to run it?

None. This is a closed model — its weights were never published, so it cannot be downloaded or run on your own hardware at any price. It is reachable only through its provider.

Source

Original publication

Record last updated 28 November 2025

The other direction

Looking at it from the other side?

This page starts from the model. If you already own a card and want to know everything it will run, start from the hardware instead.