Joint M-Best-Diverse Labelings as a Parametric Submodular Minimization

Joint M-Best-Diverse Labelings as a Parametric Submodular Minimization
Joint M-Best-Diverse Labelings as a Parametric Submodular Minimization Alexander Kirillov, Alexander Shekhovtsov, Carsten Rother, Bogdan Savchynskyy NIPS 2016 概要 Joint M-best-diverse labeling 問題をパラメトリック劣モジュラ最小化問題に帰着して解く動機づけ・問題定式化 2値画像のノイズ除去・画像分割エネルギー関数最小化として定式化エネルギー関数データ項= 「入力と出力の近さ」＋平滑化項=「出力の滑らかさ」例 $$ E(y) = \sum_{v \in V}b_v [x_v \neq y_v] + ...
論文一覧
...rnational Joint Conference on Artificial Intelligence IJCAI 2001 Link Analysis, Eigenvectors and Stability IJCAI 2009 Efficient Estimation of Influence Functions for SIS Model on Social Networks IJCAI 2011 Fast Approximate Nearest-Neighbor Search with k-Nearest Neighbor Graph IJCAI 2015 Influence Maximization in Big Networks An Incremental Algor...
@wiki全体から「Joint M-Best-Diverse Labelings as a Parametric Submodular Minimization」で調べる

todo314 @ ウィキ