Deformable DETR: Deformable Transformers for End-to-End Object Detection

DETR has been recently proposed to eliminate the need for many hand-designed components in object detection while demonstrating good performance.

249
3.08 stars / hour

deepmind-research

This repository contains implementations and illustrative code to accompany DeepMind publications

3,766
1.86 stars / hour

Is a Green Screen Really Necessary for Real-Time Portrait Matting?

24 Nov 2020ZHKKKe/MODNet

For portrait matting without the green screen, existing works either require auxiliary inputs that are costly to obtain or use multiple models that are computationally expensive.

385
1.80 stars / hour

Navigating the GAN Parameter Space for Semantic Image Editing

27 Nov 2020yandex-research/navigan

In contrast to existing works, which mostly operate by latent codes, we discover interpretable directions in the space of the generator parameters.

146
1.79 stars / hour

Fully Convolutional Networks for Panoptic Segmentation

1 Dec 2020yanwei-li/PanopticFCN

In this paper, we present a conceptually simple, strong, and efficient framework for panoptic segmentation, called Panoptic FCN.

48
1.66 stars / hour

U$^2$-Net: Going Deeper with Nested U-Structure for Salient Object Detection

18 May 2020NathanUA/U-2-Net

In this paper, we design a simple yet powerful deep network architecture, U$^2$-Net, for salient object detection (SOD).

3,308
1.17 stars / hour

End-to-End Chinese Landscape Painting Creation Using Generative Adversarial Networks

11 Nov 2020alicex2020/Chinese-Landscape-Painting-Dataset

Current GAN-based art generation methods produce unoriginal artwork due to their dependence on conditional input.

140
0.95 stars / hour

TinaFace: Strong but Simple Baseline for Face Detection

Face detection has received intensive attention in recent years.

148
0.93 stars / hour

Exploring Simple Siamese Representation Learning

20 Nov 2020sayakpaul/SimSiam-TF

Our experiments show that collapsing solutions do exist for the loss and structure, but a stop-gradient operation plays an essential role in preventing collapsing.

72
0.77 stars / hour

Sparse R-CNN: End-to-End Object Detection with Learnable Proposals

25 Nov 2020PeizeSun/SparseR-CNN

In our method, however, a fixed sparse set of learned object proposals, total length of $N$, are provided to object recognition head to perform classification and location.

587
0.68 stars / hour