Vision-based obstacle avoidance system (2005)

Closed weights New York University (NYU),Net-Scale technologies,NEC Laboratories 72K parameters December 2005

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
New York University (NYU),Net-Scale technologies,NEC Laboratories
Organisation type
Academia,Industry,Industry
Country
United States of America
Published
5 December 2005
Authors
Urs Muller, Jan Ben, Eric Cosatto, Beat Flepp, Yann LeCun

What it does

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

Domain
Robotics, Vision
Task
Self-driving car, Object detection
Approach
Supervised

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
72K

"The network has 3.15 Million connections and about 72,000 trainable parameters."

Training data
tokens

"With 95,000 training image pairs, training took 18 epochs through the training set."

Epochs
18

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 HT 630
Chips used
1
Wall-clock time
96 hours

"A complete training session required about four days of CPU time on a 3.0GHz Pentium/Xeon-based server"

Power draw
103 W

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
Training code
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
Citations
742

Sources

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

Reference
Off-Road Obstacle Avoidance through End-to-End Learning
Last updated
11 February 2026

What the numbers mean

About this model

Vision-based obstacle avoidance system (2005) was published by New York University (NYU),Net-Scale technologies,NEC Laboratories, in United States of America, in December 2005. The organisation is categorised as academia,Industry,Industry.

It works in Robotics, Vision, and is recorded as doing self-driving car, Object detection.

Because the weights are not available, none of the hardware figures elsewhere on this site apply to it.

Answers

Vision-based obstacle avoidance system (2005) — common questions

01

What is Vision-based obstacle avoidance system (2005) used for?

Vision-based obstacle avoidance system (2005) works in Robotics, Vision, and is recorded as handling self-driving car, Object 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.

02

What GPU do I need to run Vision-based obstacle avoidance system (2005)?

None. Vision-based obstacle avoidance system (2005) 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.

03

Is Vision-based obstacle avoidance system (2005) open source?

No. Vision-based obstacle avoidance system (2005) has not had its weights published, so it exists only as a service controlled by its owner.

04

How many parameters does Vision-based obstacle avoidance system (2005) have?

Vision-based obstacle avoidance system (2005) has 72K parameters. "The network has 3.15 Million connections and about 72,000 trainable parameters.". 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.

05

Who created Vision-based obstacle avoidance system (2005)?

Vision-based obstacle avoidance system (2005) was published by New York University (NYU),Net-Scale technologies,NEC Laboratories, based in United States of America, categorised as academia,Industry,Industry.

06

When was Vision-based obstacle avoidance system (2005) released?

Vision-based obstacle avoidance system (2005) was published in December 2005. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

Source

Original publication

Record last updated 11 February 2026

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.