Contour-based instance segmentation methods have developed rapidly recently but feature rough and hand-crafted front-end contour initialization, which restricts the model performance, and an empirical and fixed backen...
详细信息
Kidney tumor grade identification by means of feature based classification combined with semantic segmentation of Computed Tomography (CT) images targeting tumor region extraction are the two main contributions of thi...
详细信息
Fine grained image classification is a very popular research topic in the fields of computer vision and patternrecognition in recent years. At present, fine-grained image classification by deep learning is mainly bas...
详细信息
Currently, there are several handwriting recognition models available that effectively address the challenge of Handwritten Text recognition (HTR). However, majority of tasks require downstream training tailored for s...
详细信息
Artificial intelligence-based the Internet of Vehicles(IoV) have great significance and value to improve the driving safety of vehicles. The research mainly focuses on face recognition of drivers in the car, from thre...
详细信息
This work studies knowledge distillation (KD) and addresses its constraints for recurrent neural network transducer (RNN-T) models. In hard distillation, a teacher model transcribes large amounts of unlabelled speech ...
详细信息
Action detection aims to localize the starting and ending points of action instances in untrimmed videos, and predict the classes of those instances. In this paper, we make the observation that the outputs of the acti...
详细信息
ISBN:
(数字)9798350353006
ISBN:
(纸本)9798350353013
Action detection aims to localize the starting and ending points of action instances in untrimmed videos, and predict the classes of those instances. In this paper, we make the observation that the outputs of the action detection task can be formulated as images. Thus, from a novel perspective, we tackle action detection via a three-image generation process to generate starting point, ending point and action-class predictions as images via our proposed Action Detection image Diffusion (ADI-Diff) framework. Furthermore, since our images differ from natural images and exhibit special properties, we further explore a Discrete Action-Detection Diffusion Process and a Row-Column Transformer design to better handle their processing. Our ADI-Diff framework achieves state-of-the-art results on two widely-used datasets.
Existing methods for shadow removal in high-resolution images may not be effective due to challenges such as the time-consuming nature of training and the loss of visual data during image cropping or resizing, highlig...
Existing methods for shadow removal in high-resolution images may not be effective due to challenges such as the time-consuming nature of training and the loss of visual data during image cropping or resizing, highlighting the necessity for the development of more efficient methods. In this paper, we propose a novel Pyramid Ensemble Structure (PES) for High Resolution image Shadow Removal. Our approach takes advantage of multiple scales by constructing pyramid inputs that allow for the capturing of a wide range of shadow sizes and shapes. We then train the network in pyramid stages to enhance global information processing. Furthermore, an ensemble of different shadow removal models is employed, and the maximum value is chosen to indicate the least amount of remaining shadow in the output. Experiments on both validation and testing data sets confirm the effectiveness of our method. In the image Shadow Removal Challenge competition, our method obtained 22.36 PSNR score (1st place) and 0.70 SSIM score (2nd place) on the test sets.
3D object detection is an essential perception task in autonomous driving to understand the environments. The Bird's-Eye-View (BEV) representations have significantly improved the performance of 3D detectors with ...
详细信息
The performance of rain removal methods which are based on deep learning is largely affected by the designed models and training datasets for the image rain removal tasks. Most of current state-of-the-art focus on how...
详细信息
暂无评论