Changed conv layer to use cross-correlation rather than convolution.

dlib/dnn/cpu_dlib.cpp

@@ -1631,9 +1631,11 @@ namespace dlib
 
             // now fill in the Toeplitz output matrix for the n-th sample in data.  
             size_t cnt = 0;
-            for (long r = filter_nr-1-padding_y; r-padding_y < data.nr(); r+=stride_y)
+            const long max_r = data.nr() + padding_y-(filter_nr-1);
+            const long max_c = data.nc() + padding_x-(filter_nc-1);
+            for (long r = -padding_y; r < max_r; r+=stride_y)
             {
-                for (long c = filter_nc-1-padding_x; c-padding_x < data.nc(); c+=stride_x)
+                for (long c = -padding_x; c < max_c; c+=stride_x)
                 {
                     for (long k = 0; k < data.k(); ++k)
                     {
@@ -1642,8 +1644,8 @@ namespace dlib
                             for (long x = 0; x < filter_nc; ++x)
                             {
                                 DLIB_ASSERT(cnt < output.size(),"");
-                                long xx = c-x;
-                                long yy = r-y;
+                                long xx = c+x;
+                                long yy = r+y;
                                 if (boundary.contains(xx,yy))
                                     *t = d[(k*data.nr() + yy)*data.nc() + xx];
                                 else
@@ -1676,9 +1678,11 @@ namespace dlib
             const float* t = &output(0,0);
 
             // now fill in the Toeplitz output matrix for the n-th sample in data.  
-            for (long r = filter_nr-1-padding_y; r-padding_y < data.nr(); r+=stride_y)
+            const long max_r = data.nr() + padding_y-(filter_nr-1);
+            const long max_c = data.nc() + padding_x-(filter_nc-1);
+            for (long r = -padding_y; r < max_r; r+=stride_y)
             {
-                for (long c = filter_nc-1-padding_x; c-padding_x < data.nc(); c+=stride_x)
+                for (long c = -padding_x; c < max_c; c+=stride_x)
                 {
                     for (long k = 0; k < data.k(); ++k)
                     {
@@ -1686,8 +1690,8 @@ namespace dlib
                         {
                             for (long x = 0; x < filter_nc; ++x)
                             {
-                                long xx = c-x;
-                                long yy = r-y;
+                                long xx = c+x;
+                                long yy = r+y;
                                 if (boundary.contains(xx,yy))
                                     d[(k*data.nr() + yy)*data.nc() + xx] += *t;
                                 ++t;

dlib/dnn/cudnn_dlibapi.cpp

@@ -827,7 +827,7 @@ namespace dlib
                         stride_y,
                         stride_x,
                         1, 1, // must be 1,1
-                        CUDNN_CONVOLUTION)); // could also be CUDNN_CROSS_CORRELATION
+                        CUDNN_CROSS_CORRELATION)); // could also be CUDNN_CONVOLUTION
 
                 CHECK_CUDNN(cudnnGetConvolution2dForwardOutputDim(
                         (const cudnnConvolutionDescriptor_t)conv_handle,

dlib/dnn/layers.h

@@ -160,7 +160,7 @@ namespace dlib
 
         friend void serialize(const con_& item, std::ostream& out)
         {
-            serialize("con_3", out);
+            serialize("con_4", out);
             serialize(item.params, out);
             serialize(_num_filters, out);
             serialize(_nr, out);
@@ -186,6 +186,33 @@ namespace dlib
             long nc;
             int stride_y;
             int stride_x;
+            if (version == "con_4")
+            {
+                deserialize(item.params, in);
+                deserialize(num_filters, in);
+                deserialize(nr, in);
+                deserialize(nc, in);
+                deserialize(stride_y, in);
+                deserialize(stride_x, in);
+                deserialize(item.padding_y_, in);
+                deserialize(item.padding_x_, in);
+                deserialize(item.filters, in);
+                deserialize(item.biases, in);
+                deserialize(item.learning_rate_multiplier, in);
+                deserialize(item.weight_decay_multiplier, in);
+                deserialize(item.bias_learning_rate_multiplier, in);
+                deserialize(item.bias_weight_decay_multiplier, in);
+                if (item.padding_y_ != _padding_y) throw serialization_error("Wrong padding_y found while deserializing dlib::con_");
+                if (item.padding_x_ != _padding_x) throw serialization_error("Wrong padding_x found while deserializing dlib::con_");
+                if (num_filters != _num_filters) throw serialization_error("Wrong num_filters found while deserializing dlib::con_");
+                if (nr != _nr) throw serialization_error("Wrong nr found while deserializing dlib::con_");
+                if (nc != _nc) throw serialization_error("Wrong nc found while deserializing dlib::con_");
+                if (stride_y != _stride_y) throw serialization_error("Wrong stride_y found while deserializing dlib::con_");
+                if (stride_x != _stride_x) throw serialization_error("Wrong stride_x found while deserializing dlib::con_");
+                return;
+            }
+
+
             if (version == "con_")
             {
                 deserialize(item.params, in);
@@ -237,6 +264,20 @@ namespace dlib
                 throw serialization_error("Unexpected version '"+version+"' found while deserializing dlib::con_.");
             }
 
+
+            // now flip all the filters
+            alias_tensor at(_nr, _nc);
+            size_t off = 0;
+            for (int i = 0; i < item.filters.num_samples(); ++i)
+            {
+                for (int j = 0; j < item.filters.k(); ++j)
+                {
+                    auto temp = at(item.params,off);
+                    off += _nr*_nc;
+                    temp = flipud(fliplr(mat(temp)));
+                }
+            }
+
             if (num_filters != _num_filters) throw serialization_error("Wrong num_filters found while deserializing dlib::con_");
             if (nr != _nr) throw serialization_error("Wrong nr found while deserializing dlib::con_");
             if (nc != _nc) throw serialization_error("Wrong nc found while deserializing dlib::con_");