usersan / papers

読んだ論文のメモ置き場:主にエッジAI、高速化、FPGA実装関連など
0 stars 0 forks source link

Learning Efficient Object Detection Models with Knowledge Distillation #2

Open usersan opened 4 years ago

usersan commented 4 years ago

0. 論文

Guobin Chen, Wongun Choi, Xiang Yu, Tony Han, Manmohan Chandraker

https://papers.nips.cc/paper/6676-learning-efficient-object-detection-models-with-knowledge-distillation.pdf

他の人のメモ https://qiita.com/M_Hiro/items/431e2d63d1e9134c1038 https://taku-buntu.hateblo.jp/entry/2019/01/08/032214

1. どんなもの?

Faster R-CNNをベースに蒸留とヒント学習(中間層の出力も使う)で高速化。

2. 先行研究と比べてどこがすごい?

3. 技術や手法のキモはどこ?

4. どうやって有効だと検証した?

5. 議論はある?

6. 次に読むべき論文は?