← all datasets

SAIL-Movie Character Benchmark

Emerging
1papers using it
2020first seen

The 'SAIL-Movie Character Benchmark' is a dataset containing over 169,000 face tracks curated from 240 Hollywood movies, used to evaluate robust face clustering and character labeling in long-form content through weak labels indicating whether pairs of face tracks belong to the same or different characters.

🔎 Find this dataset

Papers using SAIL-Movie Character Benchmark (1)

SAIL-Movie Character Benchmark — datasets — learning-to-hash