IvanFang
    • Create new note
    • Create a note from template
      • Sharing URL Link copied
      • /edit
      • View mode
        • Edit mode
        • View mode
        • Book mode
        • Slide mode
        Edit mode View mode Book mode Slide mode
      • Customize slides
      • Note Permission
      • Read
        • Only me
        • Signed-in users
        • Everyone
        Only me Signed-in users Everyone
      • Write
        • Only me
        • Signed-in users
        • Everyone
        Only me Signed-in users Everyone
      • Engagement control Commenting, Suggest edit, Emoji Reply
    • Invite by email
      Invitee

      This note has no invitees

    • Publish Note

      Share your work with the world Congratulations! 🎉 Your note is out in the world Publish Note No publishing access yet

      Your note will be visible on your profile and discoverable by anyone.
      Your note is now live.
      This note is visible on your profile and discoverable online.
      Everyone on the web can find and read all notes of this public team.

      Your account was recently created. Publishing will be available soon, allowing you to share notes on your public page and in search results.

      Your team account was recently created. Publishing will be available soon, allowing you to share notes on your public page and in search results.

      Explore these features while you wait
      Complete general settings
      Bookmark and like published notes
      Write a few more notes
      Complete general settings
      Write a few more notes
      See published notes
      Unpublish note
      Please check the box to agree to the Community Guidelines.
      View profile
    • Commenting
      Permission
      Disabled Forbidden Owners Signed-in users Everyone
    • Enable
    • Permission
      • Forbidden
      • Owners
      • Signed-in users
      • Everyone
    • Suggest edit
      Permission
      Disabled Forbidden Owners Signed-in users Everyone
    • Enable
    • Permission
      • Forbidden
      • Owners
      • Signed-in users
    • Emoji Reply
    • Enable
    • Versions and GitHub Sync
    • Note settings
    • Note Insights New
    • Engagement control
    • Make a copy
    • Transfer ownership
    • Delete this note
    • Save as template
    • Insert from template
    • Import from
      • Dropbox
      • Google Drive
      • Gist
      • Clipboard
    • Export to
      • Dropbox
      • Google Drive
      • Gist
    • Download
      • Markdown
      • HTML
      • Raw HTML
Menu Note settings Note Insights Versions and GitHub Sync Sharing URL Create Help
Create Create new note Create a note from template
Menu
Options
Engagement control Make a copy Transfer ownership Delete this note
Import from
Dropbox Google Drive Gist Clipboard
Export to
Dropbox Google Drive Gist
Download
Markdown HTML Raw HTML
Back
Sharing URL Link copied
/edit
View mode
  • Edit mode
  • View mode
  • Book mode
  • Slide mode
Edit mode View mode Book mode Slide mode
Customize slides
Note Permission
Read
Only me
  • Only me
  • Signed-in users
  • Everyone
Only me Signed-in users Everyone
Write
Only me
  • Only me
  • Signed-in users
  • Everyone
Only me Signed-in users Everyone
Engagement control Commenting, Suggest edit, Emoji Reply
  • Invite by email
    Invitee

    This note has no invitees

  • Publish Note

    Share your work with the world Congratulations! 🎉 Your note is out in the world Publish Note No publishing access yet

    Your note will be visible on your profile and discoverable by anyone.
    Your note is now live.
    This note is visible on your profile and discoverable online.
    Everyone on the web can find and read all notes of this public team.

    Your account was recently created. Publishing will be available soon, allowing you to share notes on your public page and in search results.

    Your team account was recently created. Publishing will be available soon, allowing you to share notes on your public page and in search results.

    Explore these features while you wait
    Complete general settings
    Bookmark and like published notes
    Write a few more notes
    Complete general settings
    Write a few more notes
    See published notes
    Unpublish note
    Please check the box to agree to the Community Guidelines.
    View profile
    Engagement control
    Commenting
    Permission
    Disabled Forbidden Owners Signed-in users Everyone
    Enable
    Permission
    • Forbidden
    • Owners
    • Signed-in users
    • Everyone
    Suggest edit
    Permission
    Disabled Forbidden Owners Signed-in users Everyone
    Enable
    Permission
    • Forbidden
    • Owners
    • Signed-in users
    Emoji Reply
    Enable
    Import from Dropbox Google Drive Gist Clipboard
       Owned this note    Owned this note      
    Published Linked with GitHub
    • Any changes
      Be notified of any changes
    • Mention me
      Be notified of mention me
    • Unsubscribe
    ###### tags: `Paper Notes` # Faster R-CNN * 原文:[Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://arxiv.org/abs/1506.01497) * 作者:Shaoqing Ren, Kaiming He (何凱明), Ross Girshick, and Jian Sun * 時間:2015 年 ### Introduction * 現今的 object detector 都是先用 region proposal method 提出 region 後,在用 region-based CNN 去預測 bounding box 與 object class。然而,region proposal method 常常成為 object detector 的效能瓶頸。例如 R-CNN 中的 Selective Search 處理一張圖片就要花上 2 秒,與 real-time 的目標相差甚遠。 * 因此,作者提出 Region Proposal Networks (RPN) 來當作 region proposal method。RPN 的輸入為 region-based CNN 的 feature maps,輸出為 objectness scores 與 object proposals。也就是說 region-based CNN 與 region proposal method 共享同一份 feature maps。 * RPN 是一個 fully convolutional networks (FCN) 並且可以 end-to-end 的訓練。 * 有了 RPN,region proposal 的時間可以從 Selective Search 的 2 秒降到 10 毫秒。 ### Faster R-CNN * 如 Figure 2 所示,Faster R-CNN 由 RPN 與 classifier (Fast R-CNN) 兩部分組成。RPN 掃過 image 後,會產生多個 region proposals,然後在根據 region proposals 切出對應的 feature maps 的部份。最後將切出的 feature maps 丟進 Fast R-CNN 做最後的處理,提出 bounding boxes。 <center><img src="https://i.imgur.com/yedbCIi.png" width=350></center> * RPN 架構如 Figure A 所示。在這篇論文中,作者選用了 ZF-Net (與 VGG-16) 作為 RPN 中的 CNN 網路。 * image 經過 ZF-Net 後,會得到 $13 \times 13 \times 256$ 的 feature maps。 * 再將這個 feature maps 經過一層 $3 \times 3$ 的 conv. layer,一樣得到 $13 \times 13 \times 256$ 的 feature maps。 * 最後將 feature maps 分別送入兩個 $1 \times 1$ 的 conv. layer,cls. layer 與 reg. layer。 * cls. layer 用於分類該 grid 上有無 object。reg. layer 用於回歸 region proposal 的中心座標、寬高 $(x, y, w, h)$。 * cls. layer 的 filter 數為 $2K$。reg. layer 的 filter 數為 $4K$。 * $K$ 表示每個 grid 上的 anchor 數。 <center><img src="https://i.imgur.com/VEwHGap.png" width=600></center> <center>Figure A: RPN 架構圖。</center> * 總共用了 9 種 anchor box ($K = 9$): * 3 scales:128、256、512 pixels * 3 rations:(1:1)、(2:1)、(1:2) * 此外,作者還實驗了不同組合的 anchor box 的效能,如 Table 8 所示。 <center><img src="https://i.imgur.com/eHX10wq.png" width=350></center> * Loss Function of RPN: $$ L(\{p_i\}, \{t_i\}) = \frac{1}{N_{cls}} \sum_{i} L_{cls}(p_i, p_{i}^{*}) + \lambda \frac{1}{N_{reg}} \sum_{i} p_{i}^{*} L_{reg}(t_i, t_{i}^{*}) $$ * $i$:index of an anchor in a mini-batch。 * $p_i$:第 $i$ 個 anchor box 偵測該 grid 上有物件的機率。 * $p_{i}^{*}$:第 $i$ 個 anchor box 所在的 grid 上是否有物件。有則 $p_{i}^{*}=1$,否則 $p_{i}^{*}=0$。 * $t_i$:第 $i$ 個 anchor box 的 $(x, y, w, h)$ 的預測偏誤。 * $t_i^*$:第 $i$ 個 anchor box 的 $(x, y, w, h)$ 的真實偏誤。 * $N_{cls}$:mini-batch size * $N_{reg}$:number of anchor locations * $\lambda$:balancing parameter,這裡設 10。 * bounding box regression 中 4 個參數的定義如下: <center><img src="https://i.imgur.com/AErEjam.png" width=280></center> * $x, x_a, x^*$:predicted box、anchor box、ground truth box。 * $y, w, h$ 的定義同上 * $L_{reg}$ 的定義如下 (smooth L1): $$ L_{reg}(t_i, t_i^*) = R(t_i - t_i^*) = \\ smooth_{L1}(x) = \left\{ \begin{array}{rcl} 0.5 x^2,\ if\ |x| < 1 \\ |x| - 0.5,\ otherwise \end{array}\right. $$ * Faster R-CNN 的訓練方式: 1. 單獨訓練 RPN (CNN 有先預訓練在 ImageNet 上)。 2. 利用 RPN 提出的 proposal 單獨訓練 Fast R-CNN (CNN 有先預訓練在 ImageNet 上)。 3. 固定 Fast R-CNN 的 weights 訓練 RPN。 4. 固定 RPN 的 weights 訓練 Fast R-CNN。 ### Experiments & Results * Faster R-CNN 在 Pascal VOC 2007 的訓練結果如 Table 2 所示。 <center><img src="https://i.imgur.com/4rfLwt2.png" width=700></center> ### My Point of View > 雖然 Faster R-CNN 是在 2015 年提出,距今 (2020) 也有 5 年的時間了,但還是有很多人在用。例如 MCAN (2019 VQA Challange Winner)。可見 Faster R-CNN 非常重要。

    Import from clipboard

    Paste your markdown or webpage here...

    Advanced permission required

    Your current role can only read. Ask the system administrator to acquire write and comment permission.

    This team is disabled

    Sorry, this team is disabled. You can't edit this note.

    This note is locked

    Sorry, only owner can edit this note.

    Reach the limit

    Sorry, you've reached the max length this note can be.
    Please reduce the content or divide it to more notes, thank you!

    Import from Gist

    Import from Snippet

    or

    Export to Snippet

    Are you sure?

    Do you really want to delete this note?
    All users will lose their connection.

    Create a note from template

    Create a note from template

    Oops...
    This template has been removed or transferred.
    Upgrade
    All
    • All
    • Team
    No template.

    Create a template

    Upgrade

    Delete template

    Do you really want to delete this template?
    Turn this template into a regular note and keep its content, versions, and comments.

    This page need refresh

    You have an incompatible client version.
    Refresh to update.
    New version available!
    See releases notes here
    Refresh to enjoy new features.
    Your user state has changed.
    Refresh to load new user state.

    Sign in

    Forgot password
    or
    Sign in via Facebook Sign in via X(Twitter) Sign in via GitHub Sign in via Dropbox Sign in with Wallet
    Wallet ( )
    Connect another wallet

    New to HackMD? Sign up

    By signing in, you agree to our terms of service.

    Help

    • English
    • 中文
    • Français
    • Deutsch
    • 日本語
    • Español
    • Català
    • Ελληνικά
    • Português
    • italiano
    • Türkçe
    • Русский
    • Nederlands
    • hrvatski jezik
    • język polski
    • Українська
    • हिन्दी
    • svenska
    • Esperanto
    • dansk

    Documents

    Help & Tutorial

    How to use Book mode

    Slide Example

    API Docs

    Edit in VSCode

    Install browser extension

    Contacts

    Feedback

    Discord

    Send us email

    Resources

    Releases

    Pricing

    Blog

    Policy

    Terms

    Privacy

    Cheatsheet

    Syntax Example Reference
    # Header Header 基本排版
    - Unordered List
    • Unordered List
    1. Ordered List
    1. Ordered List
    - [ ] Todo List
    • Todo List
    > Blockquote
    Blockquote
    **Bold font** Bold font
    *Italics font* Italics font
    ~~Strikethrough~~ Strikethrough
    19^th^ 19th
    H~2~O H2O
    ++Inserted text++ Inserted text
    ==Marked text== Marked text
    [link text](https:// "title") Link
    ![image alt](https:// "title") Image
    `Code` Code 在筆記中貼入程式碼
    ```javascript
    var i = 0;
    ```
    var i = 0;
    :smile: :smile: Emoji list
    {%youtube youtube_id %} Externals
    $L^aT_eX$ LaTeX
    :::info
    This is a alert area.
    :::

    This is a alert area.

    Versions and GitHub Sync
    Get Full History Access

    • Edit version name
    • Delete

    revision author avatar     named on  

    More Less

    Note content is identical to the latest version.
    Compare
      Choose a version
      No search result
      Version not found
    Sign in to link this note to GitHub
    Learn more
    This note is not linked with GitHub
     

    Feedback

    Submission failed, please try again

    Thanks for your support.

    On a scale of 0-10, how likely is it that you would recommend HackMD to your friends, family or business associates?

    Please give us some advice and help us improve HackMD.

     

    Thanks for your feedback

    Remove version name

    Do you want to remove this version name and description?

    Transfer ownership

    Transfer to
      Warning: is a public team. If you transfer note to this team, everyone on the web can find and read this note.

        Link with GitHub

        Please authorize HackMD on GitHub
        • Please sign in to GitHub and install the HackMD app on your GitHub repo.
        • HackMD links with GitHub through a GitHub App. You can choose which repo to install our App.
        Learn more  Sign in to GitHub

        Push the note to GitHub Push to GitHub Pull a file from GitHub

          Authorize again
         

        Choose which file to push to

        Select repo
        Refresh Authorize more repos
        Select branch
        Select file
        Select branch
        Choose version(s) to push
        • Save a new version and push
        • Choose from existing versions
        Include title and tags
        Available push count

        Pull from GitHub

         
        File from GitHub
        File from HackMD

        GitHub Link Settings

        File linked

        Linked by
        File path
        Last synced branch
        Available push count

        Danger Zone

        Unlink
        You will no longer receive notification when GitHub file changes after unlink.

        Syncing

        Push failed

        Push successfully