v2 projector calibration working fine

src/cml/CalibrationParser.h

@@ -19,22 +19,13 @@ namespace cml
         OpticalDevice::Config& depth_cfg )
         //OpticalDevice::Config& rgb_cfg )
       {
-        //kinect:
-        //we can use kinect rgb calibration 
-        //for depth camera bc we have registration
-        //which gives the depth map 
-        //as if it were seen by the rgb camera
-        load_intrinsics( config.kinect_intrinsics_file, false, depth_cfg );
-        //set depth extrinsics
-        //to be the world origin
-        depth_cfg.extrinsics( 
-            ofVec3f( 1.,0.,0. ),
-            ofVec3f( 0.,1.,0. ),
-            ofVec3f( 0.,0.,1. ),
-            ofVec3f( 0.,0.,0. ) );
-
-        //projector:
-        load_intrinsics( config.projector_intrinsics_file, false, proj_cfg );
+        depth_cfg.name("depth camera");
+        proj_cfg.name("projector");
+
+        //use kinect rgb calibration with registration turned on 
+        load_intrinsics( config.kinect_intrinsics_file, true, depth_cfg );
+        load_intrinsics( config.projector_intrinsics_file, true, proj_cfg );
+
         load_extrinsics( config.projector_kinect_extrinsics_file, proj_cfg );
 
       };
@@ -84,19 +75,6 @@ namespace cml
 
         device.size( imageSize.width, imageSize.height );
         device.intrinsics(cx,cy,fx,fy);
-
-        ofLogNotice("cml::CalibrationParser") 
-          << "\n" << "intrinsics loaded"
-          << "\n" << "filename: " << filename
-          << "\n" << "image width: " << imageSize.width
-          << "\n" << "image height: " << imageSize.height
-          << "\n" << "principal point: " << principalPoint
-          << "\n" << "cx: " << cx
-          << "\n" << "cy: " << cy
-          << "\n" << "fx: " << fx
-          << "\n" << "fy: " << fy
-          << "\n" << "camera matrix: \n" << cameraMatrix
-          << "\n";
       };
 
       void load_extrinsics( string filename, OpticalDevice::Config& device  )
@@ -115,6 +93,11 @@ namespace cml
         if ( T.type() != CV_32FC1 ) 
           T.convertTo( T, CV_32FC1 );
 
+        T *= 1000.0; //mts->mm units
+        //transpose to get proj->camera
+        T *= -1;
+        cv::transpose(R,R);
+
         // opencv: row-major	
 
         // R x axis
@@ -138,22 +121,9 @@ namespace cml
         ofVec3f Tvec = ofVec3f(
             T.at<float>(0,0),  //tx
             T.at<float>(1,0),  //ty
-            T.at<float>(2,0)); //tz
-
-        Tvec *= 100.0; //cm->mm scale
-
-        device.extrinsics( X, Y, Z, Tvec );
-
-        ofLogNotice("cml::CalibrationParser") 
-          << "\n" << "extrinsics loaded"
-          << "\n" << "filename: " << filename
-          << "\n" << "T: \n" << T
-          << "\n" << "R: \n" << R
-          << "\n" << "T vec: " << ofToString(Tvec)
-          << "\n" << "R x axis: " << ofToString(X)
-          << "\n" << "R y axis: " << ofToString(Y)
-          << "\n" << "R z axis: " << ofToString(Z)
-          << "\n";
+            T.at<float>(2,0)); //tz 
+
+        device.extrinsics( X, Y, Z, Tvec ); 
       };
 
   };

src/cml/CamaraLucida.cpp

@@ -96,6 +96,9 @@ namespace cml
     proj = new OpticalDevice( proj_cfg );
     //rgb = new OpticalDevice( rgb_cfg );
 
+    depth->log(OF_LOG_NOTICE);
+    proj->log(OF_LOG_NOTICE);
+
     _tex_width = config.tex_width;
     _tex_height = config.tex_height;
     _depth_width = depth->width;

src/cml/Mesh.cpp

@@ -46,7 +46,7 @@ namespace cml
       to_depth( i, &xd, &yd, &idepth );
 
       float zmm = depth_pix_mm[idepth];
-      zmm = CLAMP( ( zmm < epsilon ? 5000.0 : zmm ), 0.0, 5000.0 );
+      zmm = CLAMP( ( zmm < epsilon ? depth->far_clamp : zmm ), 0.0, depth->far_clamp );
 
       float x, y;
       depth->unproject( xd, yd, zmm, &x, &y );

src/cml/OpticalDevice.cpp

@@ -10,6 +10,7 @@ namespace cml
     height = config.height;
     near = config.near;
     far = config.far;
+    far_clamp = config.far_clamp;
     cx = config.cx;
     cy = config.cy;
     fx = config.fx;

src/cml/OpticalDevice.h

@@ -1,7 +1,6 @@
 #pragma once
 
-#include "ofVectorMath.h"
-#include "ofLog.h"
+#include "ofMain.h"
 
 namespace cml
 {
@@ -15,9 +14,16 @@ namespace cml
 
           Config() 
           {
-            //mm
+            //mm units: near far
             near = 200.0;
             far = 6000.0;
+            far_clamp = 5000.0;
+
+            //init at world origin
+            X = ofVec3f( 1.,0.,0. );
+            Y = ofVec3f( 0.,1.,0. );
+            Z = ofVec3f( 0.,0.,1. );
+            T = ofVec3f( 0.,0.,0. );
           };
 
           void size( int w, int h )
@@ -46,13 +52,19 @@ namespace cml
             Y.set( _Y );
             Z.set( _Z );
             T.set( _T );
-          };
+          }; 
 
-          float near, far;
+          float near, far, far_clamp;
           int width, height;
 
           float cx, cy, fx, fy;
           ofVec3f X, Y, Z, T;
+
+          string _name;
+          void name(string n) 
+          { _name = n; };
+          string name() 
+          { return _name; };
       };
 
       struct Frustum
@@ -96,13 +108,45 @@ namespace cml
       ofVec3f& up() { return _up; };
       ofVec3f& trg() { return _trg; };
 
-      int width, height, near, far;
+      int width, height, near, far, far_clamp;
       int cx, cy, fx, fy;
 
       OpticalDevice::Config config; 
 
+      void log( ofLogLevel level )
+      {
+        Frustum& F = _frustum;
+        ofLog(level) 
+          << "cml::OpticalDevice:" << "\n" 
+          << "name: " << config.name() << "\n" 
+          << "\n"
+          << "frustum:" << "\n" 
+          << "L: " << F.left << "\n"
+          << "R: " << F.right << "\n" 
+          << "T: " << F.top << "\n"
+          << "B: " << F.bottom << "\n"
+          << "N: " << F.near << "\n"
+          << "F: " << F.far << "\n"
+          << "\n"
+          << "extrinsics:" << "\n" 
+          << "T: " << config.T << "\n"
+          << "R x axis: " << ofToString(config.X) 
+          << "\n"
+          << "R y axis: " << ofToString(config.Y) 
+          << "\n"
+          << "R z axis: " << ofToString(config.Z) 
+          << "\n"
+          << "\n";
+      };
+
     private:
 
+      Frustum _frustum; //glFrustum( ... )
+      float _KK[16]; //glMultMatrixf( KK )
+      float _RT[16]; //glMultMatrixf( RT )
+
+      ofVec3f _loc, _fwd, _up, _trg;
+
       /*
        * Intrinsics from opencv to opengl
        *
@@ -118,7 +162,7 @@ namespace cml
 
         void make_frustum( 
           OpticalDevice::Config& _cfg, 
-          OpticalDevice::Frustum& frs )
+          OpticalDevice::Frustum& F )
         {
 
           float w = _cfg.width;
@@ -130,14 +174,14 @@ namespace cml
           float far = _cfg.far;
           float near = _cfg.near;
 
-          frs.left = near * (-cx) / fx;
-          frs.right = near * (w - cx) / fx;
+          F.left = near * (-cx) / fx;
+          F.right = near * (w - cx) / fx;
 
-          frs.bottom = near * (cy - h) / fy; 
-          frs.top = near * (cy) / fy;
+          F.bottom = near * (cy - h) / fy; 
+          F.top = near * (cy) / fy;
 
-          frs.near = near;
-          frs.far = far;
+          F.near = near;
+          F.far = far; 
         };  
 
         /*
@@ -187,13 +231,7 @@ namespace cml
           RT[1]= x.y;	RT[5]= y.y;	RT[9]= z.y;	RT[13]= t.y;
           RT[2]= x.z;	RT[6]= y.z;	RT[10]=z.z; RT[14]= t.z;
           RT[3]= 0.;	RT[7]= 0.;	RT[11]= 0.;	RT[15]= 1.; 
-        };
-
-      Frustum _frustum; //glFrustum( ... )
-      float _KK[16]; //glMultMatrixf( KK )
-      float _RT[16]; //glMultMatrixf( RT )
-
-      ofVec3f _loc, _fwd, _up, _trg;
+        }; 
   }; 
 };