Home

Info Title: An Internal Learning Approach to Video Inpainting Task: Video Inpainting Author: Haotian Zhang, Long Mai, Ning Xu, Zhaowen Wang, JohnCollomosse, Hailin Jin Date: Sep. 2019 Arxiv: 1909.07957 Published: ICCV 2019 Highlights Trained on the input video(with holes) only Jointly synthesizing content in both appearan...

Color Space: RGB2BGR, RGB2GRAY import cv2 import numpy as np def iminvert(img): """Invert (negate) an image Args: img (ndarray): Image to be inverted. Returns: ndarray: The inverted image. """ return np.full_like(img, 255) - img BGR2GRAY and GRAY2BGR with OpenCV def bgr2gray(img, keepdim=False): ""...

Info Title: “Zero-Shot” Super-Resolution using Deep Internal Learning Task: Super-Resolution Author: Assaf Shocher, Nadav Cohen, Michal Irani Date: Dec. 2017 Arxiv: 1712.06087 Published: CVPR 2018 Highlights Test-time training for image-specific CNN. It can handle non-ideal imaging conditions, and a wide variet...

SRGAN: Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network - CVPR 2017 Despite the breakthroughs in accuracy and speed of single image super-resolution using faster and deeper convolutional neural networks, one central problem remains largely unsolved: how do we recover the finer texture details when we super-re...

Image Super Resolution Task Single image super-resolution (SISR) is a notoriously challenging ill-posed problem, because a specific low-resolution (LR) input can correspond to a crop of possible high-resolution (HR) images, and the HR space (in most instances it refers to the nature image space) that we intend to map the LR input to is usually ...

Traditional PatchMatch A Randomized Correspondence Algorithm for Structural Image Editing - SIGGRAPH 2009 This paper presents interactive image editing tools using a new randomized algorithm for quickly finding approximate nearest-neighbor matches between image patches. Previous research in graphics and vision has leveraged such nearest-neig...

Info Title: Imbalance Problems in Object Detection: A Review Task: task Author: Kemal Oksuz, Baris Can Cam , Sinan Kalkan, and Emre Akbas Date: Sep. 2019 Arxiv: 1909.00169 Highlights & Drawbacks We identify and define imbalance problems and present two taxonomies: A problem-based taxonomy for presenting the problems, and a s...

Anchor-based Object Detection The supervision pairs in Object Detection are produced before the conv network, using an assignment operation, in which ground truth bboxes are distributed to generated anchors based on IOU metric. The Assignment Strategy Each proposals will be assigned with -1, 0, or a positive integer indicating the ground trut...