Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileAcademic lineage
Possible advisorsa guess from early papers, not confirmed
- Bernt SchieleSuggested from co-authorship
- Cordelia SchmidSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works188 from public data
- 1,990
- Zero-Shot Learning—A Comprehensive Evaluation of the Good, the Bad and the Ugly1,908
A new zero-shot learning dataset is proposed, the Animals with Attributes 2 (AWA2) dataset which is made publicly available both in terms of image features and the images themselves and compares and analyzes a significant number of the state-of-the-art methods in depth.
- Feature Generating Networks for Zero-Shot Learning1,068
A novel generative adversarial network (GAN) that synthesizes CNN features conditioned on class-level semantic information, offering a shortcut directly from a semantic descriptor of a class to a class-conditional feature distribution.
- Zero-Shot Learning — The Good, the Bad and the Ugly987
It is argued that it is time to take a step back and to analyze the status quo of the area, given the fact that there is no agreed upon zero-shot learning benchmark, and defines a new benchmark by unifying both the evaluation protocols and data splits.
- Learning Deep Representations of Fine-Grained Visual Descriptions933
This model achieves strong performance on zero-shot text-based image retrieval and significantly outperforms the attribute-based state-of-the-art for zero- shot classification on the Caltech-UCSD Birds 200-2011 dataset.
- Label-Embedding for Image Classification871
This work proposes to view attribute-based image classification as a label-embedding problem: each class is embedded in the space of attribute vectors, and introduces a function that measures the compatibility between an image and a label embedding.
- Latent Embeddings for Zero-Shot Classification755
A novel latent embedding model for learning a compatibility function between image and class embeddings, in the context of zero-shot classification, that improves the state-of-the-art for various classembeddings consistently on three challenging publicly available datasets for the zero- shot setting.
- Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders716
This work proposes a model where a shared latent space of image features and class embeddings is learned by modality-specific aligned variational autoencoders, and align the distributions learned from images and from side-information to construct latent features that contain the essential multi-modal information associated with unseen classes.
- 711
- Generating Visual Explanations660
A new model is proposed that focuses on the discriminating properties of the visible object, jointly predicts a class label, and explains why the predicted label is appropriate for the image, and generates sentences that realize a global sentence property, such as class specificity.
- 645
- F-VAEGAN-D2: A Feature Generating Framework for Any-Shot Learning589
A conditional generative model that combines the strength of VAE and GANs and in addition, via an unconditional discriminator, learns the marginal feature distribution of unlabeled images is developed.
- Multimodal Explanations: Justifying Decisions and Pointing to the Evidence485
It is quantitatively shown that training with the textual explanations not only yields better textual justification models, but also better localizes the evidence that supports the decision, supporting the thesis that multimodal explanation models offer significant benefits over unimodal approaches.
- Textual Explanations for Self-Driving Vehicles481
A new approach to introspective explanations is proposed which uses a visual (spatial) attention model to train a convolutional network end-to-end from images to the vehicle control commands, and two approaches to attention alignment, strong- and weak-alignment are explored.
- 406
- 300
- Explainable AI: The New 42?300
Explainable AI is not a new field but the evolution of formal reasoning architectures to incorporate principled probabilistic reasoning helped address the capture and use of uncertain knowledge.
- Grounding Visual Explanations242
A phrase-critic model to refine generated candidate explanations augmented with flipped phrases to improve the textual explanation quality of fine-grained classification decisions on the CUB dataset by mentioning phrases that are grounded in the image.
- Semi-Supervised and Unsupervised Deep Visual Learning: A Survey228
This survey reviews the recent advanced deep learning algorithms on semi- supervised learning and unsupervised learning for visual recognition from a unified perspective and proposes a unified taxonomy to offer a holistic understanding of the state-of-the-art in these areas.
- Evaluating Weakly Supervised Object Localization Methods Right215
It is argued that WSOL task is ill-posed with only image-level labels, and a new evaluation protocol is proposed where full supervision is limited to only a small held-out set not overlapping with the test set.
- Exploiting Saliency for Object Segmentation from Image Level Labels211
This paper proposes using a saliency model as additional information and hereby exploit prior knowledge on the object extent and image statistics and shows how to combine both information sources in order to recover 80% of the fully supervised performance of pixel-wise semantic labelling.
- Learning Graph Embeddings for Compositional Zero-shot Learning205
A novel graph formulation called Compositional Graph Embedding (CGE) that learns image features, compositional classifiers and latent representations of visual primitives in an end-to-end manner and significantly outperforms the state of the art on MIT-States and UT-Zappos in the challenging generalized compositional zero-shot setting.
- 205
- Open World Compositional Zero-Shot Learning179
While the simple CZSL model achieves state-of-the-art performances in the closed world scenario, the feasibility scores boost the performance of the approach in the open world setting, clearly outperforming the previous state of the art.
- 176
- Semantically Tied Paired Cycle Consistency for Zero-Shot Sketch-Based Image Retrieval165
A semantically aligned paired cycle-consistent generative (SEM-PCYC) model for zero-shot SBIR, where each branch maintains a cycle consistency that only requires supervision at category levels, and avoids the need of highly-priced aligned sketch-image pairs.
- 158
- Multi-cue Zero-Shot Learning with Strong Supervision152
This work introduces a joint embedding framework that maps multiple text parts as well as multiple semantic parts into a common space that consistently and significantly improves on the state-of-the-art in zero-short recognition and retrieval.
- Waffling around for Performance: Visual Classification with Random Words and Broad Concepts138
This work critically study this behavior and proposes WaffleCLIP, a framework for zero-shot visual classification which simply replaces LLM-generated descriptors with random character and word descriptors, and demonstrates how - if available - semantic context is better leveraged by querying LLMs for high-level concepts.
- Audio Retrieval With Natural Language Queries: A Benchmark Study137
This work employs three challenging new benchmarks to establish baselines for cross-modal text-audio and audio-text retrieval, and introduces the SoundDescs benchmark, which consists of paired audio and natural language descriptions for a diverse collection of sounds that are complementary to those found in AudioCaps and Clotho.
- e-ViL: A Dataset and Benchmark for Natural Language Explanations in Vision-Language Tasks122
e-ViL is a benchmark for explainable vision-language tasks that establishes a unified evaluation framework and provides the first comprehensive comparison of existing approaches that generate NLEs for VL tasks.
- Gaze Embeddings for Zero-Shot Image Classification117
This work proposes a method that relies on human gaze as auxiliary information, exploiting that even non-expert users have a natural ability to judge class membership, and presents a data collection paradigm that involves a discrimination task to increase the information content obtained from gaze data.
- Learning Graph Embeddings for Open World Compositional Zero-Shot Learning113
This work proposes a new approach, Compositional Cosine Graph Embedding (Co-CGE), which achieves state-of-the-art performances in standard CZSL while outperforming previous methods in the open world scenario.
- How should the advancement of large language models affect the practice of science?105
Four diverse groups of scientists are invited to reflect on the advancement of large language models and their impact on both current and future scientific practices, sharing their perspectives and engaging in debate.
- Distilling Audio-Visual Knowledge by Compositional Contrastive Learning103
The main idea is to learn a compositional embedding that closes the cross-modal semantic gap and captures the task-relevant semantics, which facilitates pulling together representations across modalities by compositional contrastive learning.
- Non-negative Matrix Factorization in Multimodality Data for Segmentation and Label Prediction102
First experiments in image segmentation and multiview clustering of image features and image labels show promising results and indicate that the proposed method offers a common framework for image analysis on different levels of abstraction.
- Audio Retrieval with Natural Language Queries93
This work introduces challenging new benchmarks for text-based audio retrieval using text annotations sourced from the Audiocaps and Clotho datasets and employs these benchmarks to establish baselines for cross-modal audio retrieval, where the benefits of pre-training on diverse audio tasks are demonstrated.
- 87
- 86
- Large Loss Matters in Weakly Supervised Multi-Label Classification83
Novel methods for WSML which reject or correct the large loss samples to prevent model from memorizing the noisy label are proposed, validating that treating large loss properly matters in a weakly supervised multi-label classification.
- A Closer Look at Self-training for Zero-Label Semantic Segmentation79
This approach achieves the new state-of-the-art on PascalVOC12 and COCO-stuff datasets in the challenging generalized zero-label semantic segmentation setting, surpassing other existing methods addressing this task with more complex strategies.
- Inducing anxiety in large language models increases exploration and bias77
The lens of computational psychiatry is proposed to turn the lens of computational psychiatry, a framework used to computationally describe and modify aberrant behavior, to the outputs produced by large language models, to subject the Generative Pre-Trained Transformer 3.5 to tasks commonly studied in psychiatry.
- 75
- KG-SP: Knowledge Guided Simple Primitives for Open World Compositional Zero-Shot Learning71
The model, Knowledge-Guided Simple Primitives (KG-SP), achieves state of the art in both OW-CZ SL and pCZSL, surpassing most recent competitors even when coupled with semi-supervised learning techniques.
- Attention Consistency on Visual Corruptions for Single-Source Domain Generalization69
This work alters the content of the training images to simulate new domains and imposes consistent visual attention across the different views of the same sample, and discovers that the first objective can be simply and effectively met through visual corruptions.
- VGSE: Visually-Grounded Semantic Embeddings for Zero-Shot Learning68
This model visually divides a set of images from seen classes into clusters of local image regions according to their visual similarity, and further imposes their class discrimination and semantic relatedness, and discovers semantic embeddings that model the visual properties of both seen and unseen classes.
- Audiovisual Generalised Zero-shot Learning with Cross-modal Attention and Language64
This paper introduces a (generalised) zero-shot learning benchmark on three audio-visual datasets of varying sizes and difficulty, VGGSound, UCF, and ActivityNet, ensuring that the unseen test classes do not appear in the dataset used for supervised training of the backbone deep models.
- Exposing and Mitigating Spurious Correlations for Cross-Modal Retrieval58
This work introduces ODmAP@k, an object decorrelation metric that measures a model’s robustness to spurious correlations in the training data, and applies the proposed pipeline, which involves the finetuning of image-text retrieval frameworks on carefully designed synthetic data, to three state-of-the-art models for image- text retrieval.
- Attribute Prototype Network for Any-Shot Learning53
A novel representation learning framework that jointly learns discriminative global and local features using only class-level attributes and achieves a new state-of-the-art on challenging benchmarks, i.e. CUB, AWA2, and SUN.
- Non-isotropy Regularization for Proxy-based Deep Metric Learning52
Extensive experiments highlight consistent generalization benefits of NIR while achieving competitive and state-of-the-art performance on the standard benchmarks CUB200-2011, Cars196 and Stanford Online Products.
- 51
- 49
- Probabilistic Compositional Embeddings for Multimodal Image Retrieval48
This work proposes a novel multimodal probabilistic composer (MPC) that significantly outperforms existing related methods on multi-modal image retrieval while generalizing well to query with different amounts of inputs given in arbitrary visual and (or) textual modalities.
- Classifying the unknown: Insect identification with deep hierarchical Bayesian learning47
A novel deep hierarchical Bayesian model for insect classification that can simultaneously classify samples of both described and undescribed species, a functionality that could become instrumental in biodiversity monitoring across the globe.
- Driver Intention Anticipation Based on In-Cabin and Driving Scene Monitoring47
This work proposes a Convolutional-LSTM (ConvL STM)-based auto-encoder to extract motion features from the out-cabin traffic, trains a classifier which considers motions from both in and outside of the cabin jointly for maneuver intention anticipation, and experimentally proves that the in- and outside image features have complementary information.
- Uncertainty-Guided Progressive GANs for Medical Image Translation37
This work proposes an uncertainty-guided progressive learning scheme for image-to-image translation by incorporating aleatoric uncertainty as attention maps for GANs trained in a progressive manner, which improves performance over state-of-the-art under full-supervision and weak- supervision with limited data.
- 35
- Integrating Language Guidance into Vision-based Deep Metric Learning34
Leveraging language embeddings of expert- and pseudo-classnames, this work contextualize and realign visual representation spaces corresponding to meaningful language semantics for better semantic consistency in DML.
- 29
- 28
- 26
- 25
- 24
- 23
- 22
- 22
- 21
- 21
- 21
- 21
- 20
- 19
- 19
- 18
- 18
- 18
- 17
- 17
- 16
- 15
- DD-Ranking: Rethinking the Evaluation of Dataset Distillation14
DD-Ranking, a unified evaluation framework, along with new general evaluation metrics to uncover the true performance improvements achieved by different methods are proposed, which provide a more comprehensive and fair evaluation standard for future research advancements.
- 14
- 13
- 12
- 12
- 12
- 11
- 11
- 11
- 11
- 9
- 9
- 9
- 8
- 8
- 8
- 7
- 7
- 7
- 7
- 7
- 5
- 5
- 5
- 5
- 5
- 5
- Context-Aware Multimodal Pretraining4
It is shown that vision-language models can be trained to exhibit significantly increased few-shot adaptation, and equipped with simple, training-free, metric-based adaptation mechanisms, these representations easily surpass more complex and expensive optimization-based schemes, vastly simplifying generalization to new domains.
- 4
- 4
- 4
- 4
- 4
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-13. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.