udpate README.md

.gitignore

@@ -1,6 +1,8 @@
 build/
 rgb/
 /*.png
+/input/fountain/urd/*
+/input/fountain/gt/*
 /input/Trajectory_30_seconds/*
 !input/Trajectory_30_seconds/README.md
 /input/icl_nuim/*

README.md

@@ -1,9 +1,24 @@
-# DTAM implementation(Mapping only)
+# DTAM implementation (Mapping only)
 
-in progress.
+![image](https://user-images.githubusercontent.com/51239551/150547085-6b7eb87d-8c34-4c25-9585-48c8c7297d31.png)
+
+implementation for DTAM[Newcombe+, 2011] using C++/CUDA
 
 ## Usage
 
+### set up your environment using Docker (optional)
+follow the instruction in `docker/README.md`
+```bash
+cat docker/README.md
+```
+<br>
+
+### download `fountain-P11` dataset
+follow the instruction in `input/fountain/README.md`
+```bash
+cat input/fountain/README.md
+```
+<br>
 
 ### build
 ```bash
@@ -17,16 +32,22 @@ in progress.
 ```
 <br>
 
-## debug
+### change parameters as you like
+```bash
+vim input/json/fountain.json
+```
+<br>
+
+## debug CUDA kernel (optional)
 
-### memory checker
+### 1. memory checker
 
 ```bash
 cuda-memcheck ./build/main input/json/fountain.json
 ```
 <br>
 
-### cuda-gdb
+### 2. cuda-gdb
 
 ```bash
 cuda-gdb ./build/main

docker/README.md

@@ -45,7 +45,6 @@ export CUDA_DEBUGGER_SOFTWARE_PREEMPTION=1
 
 
 # Usage
-* update 2021,12,23
 modify `dockerfiles/Dockerfile` depending on your environment
 (such as your desired OpenCV version, Compute Capability(CC) of your GPU).
 ```bash

input/fountain/README.md

@@ -3,6 +3,9 @@ url: https://documents.epfl.ch/groups/c/cv/cvlab-unit/www/data/multiview/denseMV
 
 ## Usage
 ```bash
+cd input/fountain/
+mkdir urd
+
 curl -O https://documents.epfl.ch/groups/c/cv/cvlab-unit/www/data/multiview/data/fountain_dense/urd/fountain_dense_images.tar.gz
 tar -zxvf fountain_dense_images.tar.gz
 
@@ -12,51 +15,81 @@ tar -zxvf fountain_dense_cameras.tar.gz
 curl -O https://documents.epfl.ch/groups/c/cv/cvlab-unit/www/data/multiview/data/fountain_dense/urd/fountain_dense_p.tar.gz
 tar -zxvf fountain_dense_p.tar.gz
 
+mv fountain_dense/urd/* urd/
+
+# to generate p11.txt
 python3 gen_fou.py
 ```
 
 ## File format
+
 `urd/0000.png.camera`
 
-### K(intrinstic parameters)
+### K(intrinsic matrix)
 
-fx, 0.0,  cx,
-0.0, fy,  cy,
-0.0,0.0, 1.0
+```bash
+fx, 0, cx,
+0, fy, cy,
+0, 0, 1
+```
 
+```bash
+# line 1~3
 2759.48 0 1520.69 
 0 2764.16 1006.81 
-0 0 1 
-
-
-0 0 0
+0 0 1
+```
 
-### Rwc(Rotation matrix, from Camera to World, with respect to the World)
+### Rwc(Rotation matrix from Camera to World, with respect to the World)
 
-R00, R01, R02,
+```bash
+Rwc = 
+[R00, R01, R02,
 R10, R11, R12,
-R20, R21, R22
+R20, R21, R22]
+```
 
+```bash
+# line 5~7
 0.450927 -0.0945642 -0.887537 
 -0.892535 -0.0401974 -0.449183 
 0.00679989 0.994707 -0.102528 
+```
 
-### twc(translation, the location of the optic center of camera in the world frame)
 
-tx, ty, tz
 
+### twc(translation vector from Camera to World, the location of the optic center of camera in the world frame)
+
+```bash
+twc = 
+[tx, ty, tz]
+```
+
+```bash
+# line 8
 -7.28137 -7.57667 0.204446 
+```
+
 
-### width height
+### width and height of `urd/0000.png`
 
+```bash
+width height
+```
+
+```bash
+# line 9
 3072 2048
+```
+
 
 <br>
 <br>
 
 
 `urd/0000.png.P`
 
+### perspective projection matrix P
 ```bash
-P = K[Rcw|tcw] = K[Rwc^(-1)|-Rwc^(-1)*twc] = K[Rwc^t|-Rwc^t*twc]
+P = K[Rcw|tcw] = K[Rwc^(-1)| -Rwc^(-1)*twc] = K[Rwc^t| -Rwc^t*twc]
 ```