Trending Research
End-to-End Object Detection with Transformers
We present a new method that views object detection as a direct set prediction problem.
#2 best model for
Panoptic Segmentation
on COCO panoptic
Big Transfer (BiT): General Visual Representation Learning
• google-research/big_transfer
• 
We conduct detailed analysis of the main components that lead to high transfer performance.
SOTA for Image Classification on ObjectNet (Bounding Box) (using extra training data)
FEW-SHOT LEARNING FINE-GRAINED IMAGE CLASSIFICATION REPRESENTATION LEARNING
Point2Mesh: A Self-Prior for Deformable Meshes
We optimize the network weights to deform an initial mesh to shrink-wrap a single input point cloud.
Surfboard: Audio Feature Extraction for Modern Machine Learning
We introduce Surfboard, an open-source Python library for extracting audio features with application to the medical domain.
Instance-aware Image Colorization
Previous methods leverage the deep neural network to map input grayscale images to plausible color outputs directly.
PyContrast
PyTorch implementation of Contrastive Learning methods
CONTRASTIVE LEARNING DATA AUGMENTATION INSTANCE SEGMENTATION OBJECT DETECTION REPRESENTATION LEARNING SELF-SUPERVISED IMAGE CLASSIFICATION SEMANTIC SEGMENTATION
Skeleton-Aware Networks for Deep Motion Retargeting
• DeepMotionEditing/deep-motion-editing
• 
In other words, our operators form the building blocks of a new deep motion processing framework that embeds the motion into a common latent space, shared by a collection of homeomorphic skeletons.
Unpaired Motion Style Transfer from Video to Animation
• DeepMotionEditing/deep-motion-editing
•
In this paper, we present a novel data-driven framework for motion style transfer, which learns from an unpaired collection of motions with style labels, and enables transferring motion styles not observed during training.
GECToR -- Grammatical Error Correction: Tag, Not Rewrite
• grammarly/gector
•
In this paper, we present a simple and efficient GEC sequence tagger using a Transformer encoder.
ResNeSt: Split-Attention Networks
While image classification models have recently continued to advance, most downstream applications such as object detection and semantic segmentation still employ ResNet variants as the backbone network due to their simple and modular structure.
SOTA for Semantic Segmentation on ADE20K
IMAGE CLASSIFICATION INSTANCE SEGMENTATION OBJECT DETECTION PANOPTIC SEGMENTATION