Multiresolution CNN

Closed weights Google,Stanford University 126.1M parameters June 2014

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
Google,Stanford University
Organisation type
Industry,Academia
Country
United States of America
Published
23 June 2014
Authors
A Karpathy, G Toderici, S Shetty, T Leung

What it does

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

Domain
Video
Task
Video classification

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
126.1M

"Using shorthand notation, the full [single frame] architecture is C(96, 11, 3)-N-P-C(256, 5, 1)-N-P-C(384, 3, 1)-C(384, 3, 1)-C(256, 3, 1)-P-FC(4096)-FC(4096), where C(d, f, s) indicates a convolutional layer with d filters of spatial size f ×f, applied to the input with stride s" Two such single-frame architectures are concatenated as shown in figure 2 "Since the input is only of half the spatial size as the full-frame models, we take out the last pooling layer to ensure that both streams st…

Training data
50,000,000 tokens

"We further estimate the size of our dataset of sampled frames to be on the order of 50 million examples and that our networks have each seen approximately 500 million examples throughout the training period in total." So 5e+7 datapoints and 10 epochs.

How it is classified

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

Why it is tracked
Highly cited
Citations
6,254

Sources

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

Reference
Large-Scale Video Classification with Convolutional Neural Networks
Last updated
28 November 2025

What the numbers mean

Background

Multiresolution CNN was published by Google,Stanford University, in United States of America, in June 2014. It comes out of industry,Academia.

It works in Video, and is recorded as doing video classification.

Its weights were never published, so it can only be reached through its provider. No graphics card changes that.

How it was trained

The training set ran to roughly 50,000,000 tokens.

Its inclusion criterion is highly cited.

Answers

Multiresolution CNN — common questions

01

Is Multiresolution CNN open source?

The licensing for Multiresolution CNN was never recorded in our source data. We treat unstated licensing as closed, because an unrecorded licence is not one to rely on.

02

How many parameters does Multiresolution CNN have?

Multiresolution CNN has 126.1M parameters. "Using shorthand notation, the full [single frame] architecture is C(96, 11, 3)-N-P-C(256, 5, 1)-N-P-C(384, 3, 1)-C(384, 3, 1)-C(256, 3, 1)-P-FC(4096)-FC(4096), where C(d, f, s) indicates a convolutional layer with d filters of spatial size f ×f, applied to the input with stride s" Two such single-frame architectures are concatenated as shown in figure 2 "Since the input is only of half the spatial size as the full-frame models, we take out the last pooling layer to ensure that both streams still terminate in a layer of size 7×7×256. " We assume the input are T=10 frames with C=3 color channels each 2*(256*(10*3*5*5+1) + 384*(256*3*3+1) + 384*(384*3*3+1) + 256*(384*3*3+1)) + (2*7*7*256 + 1)*4096 + (4096+1)*4096. 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

Who created Multiresolution CNN?

Multiresolution CNN was published by Google,Stanford University, based in United States of America, categorised as industry,Academia.

04

When was Multiresolution CNN released?

Multiresolution CNN was published in June 2014. 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

What is Multiresolution CNN used for?

Multiresolution CNN works in Video, and is recorded as handling video classification. Models frequently carry more than one of each, and the tags describe purpose rather than capability limits.

06

What GPU do I need to run Multiresolution CNN?

None. Multiresolution CNN 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.