davisking/dlib

3D convolution and pooling

Open

#989 opened on Nov 30, 2017

View on GitHub
 (7 comments) (1 reaction) (0 assignees)C++ (3,314 forks)batch import
help wanted

Repository metrics

Stars
 (12,570 stars)
PR merge metrics
 (Avg merge 3h 56m) (2 merged PRs in 30d)

Description

I am currently looking for CNN training tools for windows with c++. dlib seems to be a good choice as the example clearly illustrate how ResNet is implemented.

However, I am looking for applications for 3D image segmentation which requires 3D pooling and convolution (somehow it is called 4D convolution in this link

This function is supported in caffe2, tensorflow and pytorch. Any plan to implement for a higher dimension CNN operations?

Contributor guide

Research direction
Examine the existing 2D convolution and pooling implementations in dlib (e.g., in dlib/dnn/). Understand the tensor layout and how spatial dimensions are handled. Then design an extension to support 3D (or 4D) convolutions. Review the discussion link for prior context.
Tech stack
cpp
Domain
machine learningai
Issue type
Feature
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
4
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
Over 1 week
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Mostly clear
Prerequisites
C++deep learningdlib
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
20