We argue that using a single embedding vector to represent an image, as commonly practiced, is not sufficient to rank both relevant seen and unseen labels accurately.
Ranked #2 on Multi-label zero-shot learning on Open Images V4
In this paper, we introduce a novel asymmetric loss ("ASL"), which operates differently on positive and negative samples.
Ranked #3 on Multi-Label Classification on NUS-WIDE
In this work, we introduce a series of architecture modifications that aim to boost neural networks' accuracy, while retaining their GPU training and inference efficiency.
Ranked #4 on Fine-Grained Image Classification on Oxford 102 Flowers (using extra training data)
Through this we produce compact architectures with the same FLOPs as EfficientNet-B0 and MobileNetV3 but with higher accuracy, by $1\%$ and $0. 3\%$ respectively on ImageNet, and faster runtime on GPU.
Ranked #1 on Network Pruning on ImageNet
Furthermore, we show the representation power of our ReID network via SotA results on a different task of multi-object tracking.
Ranked #9 on Person Re-Identification on Market-1501 (Rank-1 metric)
This paper introduces a novel optimization method for differential neural architecture search, based on the theory of prediction with expert advice.
In this paper, we propose a differentiable search space that allows the annealing of architecture weights, while gradually pruning inferior operations.
We present an approach to semi-supervised video object segmentation, in the context of the DAVIS 2017 challenge.