Generalized Zero-Shot Learning
55 papers with code • 12 benchmarks • 10 datasets
In generalized zero shot learning (GZSL), the set of classes are split into seen and unseen classes, where training relies on the semantic features of the seen and unseen classes and the visual representations of only the seen classes, while testing uses the visual representations of the seen and unseen classes.
Benchmarks
These leaderboards are used to track progress in Generalized Zero-Shot Learning
| Trend | Dataset | Best Model | Paper | Code | Compare |
|---|---|---|---|---|---|
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
Datasets
CUB-200-2011
ScanNet
Oxford 102 Flower
SemanticKITTI
S3DIS
AwA
OntoNotes 5.0
AwA2
aPY
SUN Attribute
Latest papers
Audio-Visual Generalized Zero-Shot Learning using Pre-Trained Large Multi-Modal Models
However, existing benchmarks predate the popularization of large multi-modal models, such as CLIP and CLAP.
Less but Better: Enabling Generalized Zero-shot Learning Towards Unseen Domains by Intrinsic Learning from Redundant LLM Semantics
Different from existing GZSL methods which alleviate DSP by generating features of unseen classes with semantics, CDGZSL needs to construct a common feature space across domains and acquire the corresponding intrinsic semantics shared among domains to transfer from seen to unseen domains.
Data Distribution Distilled Generative Model for Generalized Zero-Shot Recognition
pjbq/d3gzsl
•
•
To counter this, we introduce an end-to-end generative GZSL framework called D$^3$GZSL.
Dual Feature Augmentation Network for Generalized Zero-shot Learning
sion1/dfan
•
•
To address these issues, we propose a novel Dual Feature Augmentation Network (DFAN), which comprises two feature augmentation modules, one for visual features and the other for semantic features.
Zero-Shot Learning by Harnessing Adversarial Samples
uqzhichen/haszsl
•
•
To take the advantage of image augmentations while mitigating the semantic distortion issue, we propose a novel ZSL approach by Harnessing Adversarial Samples (HAS).
Improving Zero-Shot Generalization for CLIP with Synthesized Prompts
Thereafter, we fine-tune CLIP with off-the-shelf methods by combining labeled and synthesized features.
Progressive Semantic-Visual Mutual Adaption for Generalized Zero-Shot Learning
manliucoder/psvma
•
•
CVPR 2023
Generalized Zero-Shot Learning (GZSL) identifies unseen categories by knowledge transferred from the seen domain, relying on the intrinsic interactions between visual and semantic information.
Learning Generalized Zero-Shot Learners for Open-Domain Image Geolocalization
Image geolocalization is the challenging task of predicting the geographic coordinates of origin for a given photo.
On the Transferability of Visual Features in Generalized Zero-Shot Learning
Generalized Zero-Shot Learning (GZSL) aims to train a classifier that can generalize to unseen classes, using a set of attributes as auxiliary information, and the visual features extracted from a pre-trained convolutional neural network.
PROTOtypical Logic Tensor Networks (PROTO-LTN) for Zero Shot Learning
francescomanigrass/proto-ltn
•
•
We focus here on the subsumption or \texttt{isOfClass} predicate, which is fundamental to encode most semantic image interpretation tasks.
