|
1 | 1 | # Computer-Vision-Assignments |
2 | 2 |
|
3 | | -> Reports of all the assignments are available in ./Report directory. |
| 3 | +## Reports of all the assignments are available in ./Report directory. |
4 | 4 |
|
5 | | -============================================================================ |
6 | 5 | ## Assignment-1 |
7 | 6 | - Average Filter (3x3, 5x5, 11x11, and 15x15). Analysis of using avg filter with different kernel sizes. |
8 | 7 | - Adding *Salt and Pepper noise*. Removing noise using median filter with different kernel sizes. |
|
11 | 10 | - Performed **Discrete Wavelet Transform using Haar classifier** to remove High frequency component to smoothen the images. |
12 | 11 | - Performed Watermarking using DWT. |
13 | 12 |
|
14 | | -============================================================================ |
15 | 13 | ## Assignment-2 |
16 | 14 | - Implemented **Hough Tranform for circles** from scratch. |
17 | 15 | - Camera Calibration using checkerboard. |
18 | 16 | - Implemented **Harris Corner Detection** from scratch. |
19 | 17 |
|
20 | | -============================================================================ |
21 | 18 | ## Assignment-3 |
22 | 19 | - Image segmentation using **K-Means clustering** for: |
23 | 20 | - 3-D Color Space |
|
28 | 25 | - Seeded Segmentation - Manually marking point in an image and recursively analyzing nearby points. |
29 | 26 | - **Image Classification** using **Bag-of-Visual-Word** with HOG and LBP features on **CIFAR-10** Dataset. |
30 | 27 |
|
31 | | -============================================================================ |
32 | 28 | ## Assignment-4 |
33 | 29 | - **Image Classification**: Extracted features from pretrained **AlexNet** model and classification using Random Forest Classifier. |
34 | 30 | - **Convolutional Neural Network**: Created our own architecture for CNN and performed 4 class classfication from CIFAR-10 Dataset. |
35 | 31 | - **Image Identification and Matching**: Matched using **Brute-Force** feature matching technique and plotted box for matched object in images. |
36 | 32 | - **Panaroma Stiching**: Stiched 3 Taj Mahal Images into 1. |
37 | 33 |
|
38 | | -============================================================================ |
39 | 34 | ## Assignment-5 [Bonus] |
40 | 35 | - Computed **Depth Map** in given 2 images. |
41 | | - |
42 | | -============================================================================ |
0 commit comments