Hierarchical Scene Labeling (Stanford Background)

Training compute

2.4×10¹⁷ FLOP

Parameters

51.6M

Published

Aug 1, 2013

Hierarchical Scene Labeling (Stanford Background) is an AI model developed by New York University (NYU) (United States), first published in August 2013. It works in the vision domain, on tasks such as semantic segmentation.

Training it took an estimated 2.4×10¹⁷ FLOP of compute (estimation method: operation counting). The model has 51,609,600 parameters. It was trained on roughly 71.4M datapoints.

Access: Unreleased. Its weights are not openly released. Epoch AI rates the confidence of this record as confident.

Full record

Organization: New York University (NYU)
Country of organization: United States
Domain: Vision
Task: Semantic segmentation
Training compute: 2.4×10¹⁷ FLOP
Compute estimation method: Operation counting
Parameters: 51,609,600
Dataset size: 71.4M
Chips used: 1
Model accessibility: Unreleased
Open weights: No
Epoch confidence: Confident