Then, we demonstrate how a trained landmark detector, using our new dataset, can be leveraged to index image regions and improve retrieval accuracy while being much more efficient than existing regional methods.
Many of the recent successful methods for video object segmentation (VOS) are overly complicated, heavily rely on fine-tuning on the first frame, and/or are slow, and are hence of limited practical use.
We propose an alternative generator architecture for generative adversarial networks, borrowing from style transfer literature.
In this work, we carefully revisit the standard training practice of detectors, and find that the detection performance is often limited by the imbalance during the training process, which generally consists in three levels - sample level, feature level, and objective level.
#21 best model for Object Detection on COCO test-dev
In this paper, we study this problem and propose Mask Scoring R-CNN which contains a network block to learn the quality of the predicted instance masks.
#5 best model for Instance Segmentation on COCO
The superior performance of Deformable Convolutional Networks arises from its ability to adapt to the geometric variations of objects.
#9 best model for Object Detection on COCO test-dev
In exploring a more effective approach, we find that the key to a successful instance segmentation cascade is to fully leverage the reciprocal relationship between detection and segmentation.
SOTA for Instance Segmentation on COCO (using extra training data)
We propose spatially-adaptive normalization, a simple but effective layer for synthesizing photorealistic images given an input semantic layout.