With the continuous improvement of computing power and deep learning
alg...
The "pre-training → downstream adaptation" presents both new
opportuniti...
Multi-scale features are essential for dense prediction tasks, including...
Knowledge distillation is an effective method for model compression. How...
DETR is a novel end-to-end transformer architecture object detector, whi...
We present a strong object detector with encoder-decoder pretraining and...
The requirement of expensive annotations is a major burden for training ...
This paper proposes a novel Unified Feature Optimization (UFO) paradigm ...
LiDAR semantic segmentation essential for advanced autonomous driving is...
Extracting cultivated land accurately from high-resolution remote images...
Despite the recent success of long-tailed object detection, almost all
l...
Despite superior performance on many computer vision tasks, deep convolu...
Microservice architecture advocates a number of technologies and practic...
Learning discriminative representation using large-scale face datasets i...
The two-stage methods for instance segmentation, e.g. Mask R-CNN, have
a...
Recently proposed decoupled training methods emerge as a dominant paradi...
Modern deep neural network models are large and computationally intensiv...
This paper introduces the real image Super-Resolution (SR) challenge tha...
This article introduces the solutions of the team lvisTraveler for LVIS
...
With advancement in deep neural network (DNN), recent state-of-the-art (...
This paper reviews the NTIRE 2020 challenge on real image denoising with...