Optical Flow 發展歷程 (1)

Optical flow estimation

Traditional Method

Variational approach

• • TVL-1 [1]

Deep Method

Supervised

• • FlowNet [6]
  • FlowNetC [6]
  • FlowNetS [6]
  • FlowNet2.0 [3]
  • SpyNet [5]
  • LiteFlowNet [7]
  • PWCNet [4]

Unsupervised

• • MotionFlowNet [2]

 

 

接下來的幾篇文章將詳細分析這幾個模型：

 

---

 

參考文獻：

[1] A Duality Based Approach for Realtime TV-L1 Optical Flow

[2] Hidden Two-Stream Convolutional Networks for Action Recognition

[3] Flownet 2.0: Evolution of optical flow estimation with deep networks

[4] PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volume

[5] Optical flow estimation using a spatial pyramid network

[6] Flownet: Learning optical flow with convolutional networks

[7] LiteFlowNet: A Lightweight Convolutional Neural Network for Optical Flow Estimation