Fig. 1. Reshapeschematic of SIFT feature descriptor

Fig. 2. SIFT feature descriptors after changing shape. (a) Feature descriptor; (b) feature descriptor matched with Fig. 2(a); (c) feature descriptor unmatched with Fig. 2(a)

Fig. 3. Diagram of convolutional neural network for SIFT feature descriptor dimensionality reduction

Fig. 4. Partial atlases used to test matching performance of feature descriptor

Fig. 5. Comparison of partial matching results between CNN-SIFT feature descriptor and SIFT feature descriptor in affine transformation

Fig. 6. Matching performance of feature descriptors in rotation and scale transformations

Fig. 7. Matching performance of feature descriptors in viewpoint transformation

Fig. 8. Matching performance of feature descriptors in light transformation

Fig. 9. Average matching time of feature descriptors in Webcam dataset