@@ -172,16 +172,19 @@ <h2 class="title is-3" style="white-space: nowrap;">Reconstruct images from feat
172172 < div class ="column is-four-fifths ">
173173 < h2 class ="title is-3 " style ="white-space: nowrap; "> Feature-space transformations. Q matrix Calculation and Application.</ h2 >
174174 < div class ="content has-text-justified ">
175- < div style ="text-align: center; ">
176- < img src ="./static/images/features_reconstruction_manipulation_train_Q.drawio.png " alt ="features_reconstruction_manipulation_train_Q " width ="70% ">
177- < p class ="caption " style ="width: 70%; text-align: center; ">
178- < b > Figure 3. Feature-space transformations. Q matrix Calculation.</ b > We then calculate Q matrix for feature-space manupulation.
179- </ p >
180- < br > < br >
181- < img src ="./static/images/features_reconstruction_manipulation_eval_Q.drawio.png " alt ="features_reconstruction_manipulation_eval_Q " width ="70% ">
182- < p class ="caption " style ="width: 70%; text-align: center; ">
183- < b > Figure 4. Feature-space transformations. Q matrix Application.</ b > After Q matrix is calculated, we apply it to the feature space. For each patch embedding.
184- </ p >
175+ < div class ="columns is-centered ">
176+ < div class ="column is-half ">
177+ < img src ="./static/images/features_reconstruction_manipulation_train_Q.drawio.png " alt ="features_reconstruction_manipulation_train_Q " width ="100% ">
178+ < p class ="caption " style ="width: 70%; text-align: center; ">
179+ < b > Figure 3. Feature-space transformations. Q matrix Calculation.</ b > We then calculate Q matrix for feature-space manupulation.
180+ </ p >
181+ </ div >
182+ < div class ="column is-half ">
183+ < img src ="./static/images/features_reconstruction_manipulation_eval_Q.drawio.png " alt ="features_reconstruction_manipulation_eval_Q " width ="100% ">
184+ < p class ="caption " style ="width: 70%; text-align: center; ">
185+ < b > Figure 4. Feature-space transformations. Q matrix Application.</ b > After Q matrix is calculated, we apply it to the feature space. For each patch embedding.
186+ </ p >
187+ </ div >
185188 </ div >
186189 </ div >
187190 < br > < br >
@@ -192,16 +195,19 @@ <h2 class="title is-3" style="white-space: nowrap;">Feature-space transformation
192195 < div class ="column is-four-fifths ">
193196 < h2 class ="title is-3 " style ="white-space: nowrap; "> Feature-space transformations. Color Swap Examples.</ h2 >
194197 < div class ="content has-text-justified ">
195- < div style ="text-align: center; ">
196- < img src ="./static/images/rb_swap.png " alt ="rb_swap " width ="70% ">
197- < p class ="caption " style ="width: 70%; text-align: center; ">
198- < b > Figure 5. Red-blue channel swap samples.</ b >
199- </ p >
200- < br > < br >
201- < img src ="./static/images/color_swap_all_eigen_values.png " alt ="color_swap_all_eigen_values " width ="70% ">
202- < p class ="caption " style ="width: 70%; text-align: center; ">
203- < b > Figure 6. Eigenvalues for red-blue channel swap matrix.</ b > Majority of eigenvalues are close to 1, which means that the transformation is close to an identity matrix. While the other cluster of eigenvalues are close to -1, which means that for these channels direction is changed to the opposite.
204- </ p >
198+ < div class ="columns is-centered ">
199+ < div class ="column is-half ">
200+ < img src ="./static/images/rb_swap.png " alt ="rb_swap " width ="100% ">
201+ < p class ="caption " style ="width: 100%; text-align: center; ">
202+ < b > Figure 5. Red-blue channel swap samples.</ b >
203+ </ p >
204+ </ div >
205+ < div class ="column is-half ">
206+ < img src ="./static/images/color_swap_all_eigen_values.png " alt ="color_swap_all_eigen_values " width ="100% ">
207+ < p class ="caption " style ="width: 100%; text-align: center; ">
208+ < b > Figure 6. Eigenvalues for red-blue channel swap matrix.</ b > Majority of eigenvalues are close to 1, which means that the transformation is close to an identity matrix. While the other cluster of eigenvalues are close to -1, which means that for these channels direction is changed to the opposite.
209+ </ p >
210+ </ div >
205211 </ div >
206212 </ div >
207213 < br > < br >
@@ -212,16 +218,19 @@ <h2 class="title is-3" style="white-space: nowrap;">Feature-space transformation
212218 < div class ="column is-four-fifths ">
213219 < h2 class ="title is-3 " style ="white-space: nowrap; "> Feature-space transformations. Blue Channel Suppression.</ h2 >
214220 < div class ="content has-text-justified ">
215- < div style ="text-align: center; ">
216- < img src ="./static/images/b_suppression_all_transformations.png " alt ="b_suppression_all_transformations " width ="70% ">
217- < p class ="caption " style ="width: 70%; text-align: center; ">
218- < b > Figure 7. Blue channel suppression samples.</ b >
219- </ p >
220- < br > < br >
221- < img src ="./static/images/b_suppression_all_eigen_values.png " alt ="b_suppression_all_eigen_values " width ="70% ">
222- < p class ="caption " style ="width: 70%; text-align: center; ">
223- < b > Figure 8. Eigenvalues for blue channel suppression matrix.</ b >
224- </ p >
221+ < div class ="columns is-centered ">
222+ < div class ="column is-half ">
223+ < img src ="./static/images/b_suppression_all_transformations.png " alt ="b_suppression_all_transformations " width ="100% ">
224+ < p class ="caption " style ="width: 100%; text-align: center; ">
225+ < b > Figure 7. Blue channel suppression samples.</ b >
226+ </ p >
227+ </ div >
228+ < div class ="column is-half ">
229+ < img src ="./static/images/b_suppression_all_eigen_values.png " alt ="b_suppression_all_eigen_values " width ="100% ">
230+ < p class ="caption " style ="width: 100%; text-align: center; ">
231+ < b > Figure 8. Eigenvalues for blue channel suppression matrix.</ b >
232+ </ p >
233+ </ div >
225234 </ div >
226235 </ div >
227236 < br > < br >
0 commit comments