Visual Question Answering On Clevr Humans

Accuracy

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

		Paper Title
MDETR	81.7	MDETR -- Modulated Detection for End-to-End Multi-Modal Understanding
MAC	81.5	Compositional Attention Networks for Machine Reasoning
CNN+GRU+FiLM	75.9	FiLM: Visual Reasoning with a General Conditioning Layer
NS-VQA (1K programs)	67.8	Neural-Symbolic VQA: Disentangling Reasoning from Vision and Language Understanding
IEP-18K	66.6	Inferring and Executing Programs for Visual Reasoning

0 of 5 row(s) selected.