Search Results for author:
Found 3 papers, 3 papers with code
FPGAN-Control: A Controllable Fingerprint Generator for Training with Synthetic Data
We demonstrate the merits of FPGAN-Control, both quantitatively and qualitatively, in terms of identity preservation level, degree of appearance control, and low synthetic-to-real domain gap.
Semantic Diversity Learning for Zero-Shot Multi-label Classification
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 #3 on
Multi-label zero-shot learning
on Open Images V4
TResNet: High Performance GPU-Dedicated Architecture
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 #7 on
Fine-Grained Image Classification
on Oxford 102 Flowers
(using extra training data)
