Search Results for author: Pinar Yanardag

Found 20 papers, 6 papers with code

GANTASTIC: GAN-based Transfer of Interpretable Directions for Disentangled Image Editing in Text-to-Image Diffusion Models

no code implementations28 Mar 2024 Yusuf Dalva, Hidir Yesiltepe, Pinar Yanardag

The rapid advancement in image generation models has predominantly been driven by diffusion models, which have demonstrated unparalleled success in generating high-fidelity, diverse images from textual prompts.

Image Generation

MIST: Mitigating Intersectional Bias with Disentangled Cross-Attention Editing in Text-to-Image Diffusion Models

no code implementations28 Mar 2024 Hidir Yesiltepe, Kiymet Akdemir, Pinar Yanardag

Addressing intersectional bias is crucial because it amplifies the negative effects of discrimination based on race, gender, and other identities.


CLoRA: A Contrastive Approach to Compose Multiple LoRA Models

no code implementations28 Mar 2024 Tuna Han Salih Meral, Enis Simsar, Federico Tombari, Pinar Yanardag

Low-Rank Adaptations (LoRAs) have emerged as a powerful and popular technique in the field of image generation, offering a highly effective way to adapt and refine pre-trained deep learning models for specific tasks without the need for comprehensive retraining.

Image Generation

CONFORM: Contrast is All You Need For High-Fidelity Text-to-Image Diffusion Models

no code implementations11 Dec 2023 Tuna Han Salih Meral, Enis Simsar, Federico Tombari, Pinar Yanardag

Images produced by text-to-image diffusion models might not always faithfully represent the semantic intent of the provided text prompt, where the model might overlook or entirely fail to produce certain objects.

NoiseCLR: A Contrastive Learning Approach for Unsupervised Discovery of Interpretable Directions in Diffusion Models

no code implementations8 Dec 2023 Yusuf Dalva, Pinar Yanardag

Our extensive experiments show that our method achieves highly disentangled edits, outperforming existing approaches in both diffusion-based and GAN-based latent space editing methods.

Contrastive Learning Image Generation

3D-LatentMapper: View Agnostic Single-View Reconstruction of 3D Shapes

no code implementations5 Dec 2022 Alara Dirik, Pinar Yanardag

Computer graphics, 3D computer vision and robotics communities have produced multiple approaches to represent and generate 3D shapes, as well as a vast number of use cases.

Fantastic Style Channels and Where to Find Them: A Submodular Framework for Discovering Diverse Directions in GANs

no code implementations16 Mar 2022 Enis Simsar, Umut Kocasari, Ezgi Gülperi Er, Pinar Yanardag

We evaluate our framework with qualitative and quantitative experiments and show that our method finds more diverse and disentangled directions.

Image Generation

Discovering Multiple and Diverse Directions for Cognitive Image Properties

no code implementations23 Feb 2022 Umut Kocasari, Alperen Bag, Oguz Kaan Yuksel, Pinar Yanardag

While previous work has focused on discovering a single direction that performs a desired editing operation such as zoom-in, limited work has been done on the discovery of multiple and diverse directions that can achieve the desired edit.

FairStyle: Debiasing StyleGAN2 with Style Channel Manipulations

no code implementations13 Feb 2022 Cemre Karakas, Alara Dirik, Eylul Yalcinkaya, Pinar Yanardag

Our experiments show that our method successfully debiases the GAN model within a few minutes without compromising the quality of the generated images.

Text and Image Guided 3D Avatar Generation and Manipulation

1 code implementation12 Feb 2022 Zehranaz Canfes, M. Furkan Atasoy, Alara Dirik, Pinar Yanardag

In this work, we propose a novel 3D manipulation method that can manipulate both the shape and texture of the model using text or image-based prompts such as 'a young face' or 'a surprised face'.


StyleMC: Multi-Channel Based Fast Text-Guided Image Generation and Manipulation

no code implementations15 Dec 2021 Umut Kocasari, Alara Dirik, Mert Tiftikci, Pinar Yanardag

Discovering meaningful directions in the latent space of GANs to manipulate semantic attributes typically requires large amounts of labeled data.

Image Generation Prompt Engineering

Exploring Latent Dimensions of Crowd-sourced Creativity

1 code implementation13 Dec 2021 Umut Kocasari, Alperen Bag, Efehan Atici, Pinar Yanardag

We explore the latent dimensions of images generated on this platform and present a novel framework for manipulating images to make them more creative.

Controlled Cue Generation for Play Scripts

no code implementations13 Dec 2021 Alara Dirik, Hilal Donmez, Pinar Yanardag

In this paper, we use a large-scale play scripts dataset to propose the novel task of theatrical cue generation from dialogues.

Attribute Language Modelling +1

LatentCLR: A Contrastive Learning Approach for Unsupervised Discovery of Interpretable Directions

2 code implementations ICCV 2021 Oğuz Kaan Yüksel, Enis Simsar, Ezgi Gülperi Er, Pinar Yanardag

Recent research has shown that it is possible to find interpretable directions in the latent spaces of pre-trained Generative Adversarial Networks (GANs).

Contrastive Learning Image Generation

A Structural Smoothing Framework For Robust Graph Comparison

no code implementations NeurIPS 2015 Pinar Yanardag, S. V. N. Vishwanathan

In this paper, we propose a general smoothing framework for graph kernels by taking \textit{structural similarity} into account, and apply it to derive smoothed variants of popular graph kernels.

Deep Graph Kernels

no code implementations KDD '15 Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining 2015 Pinar Yanardag, S. V. N. Vishwanathan

In this paper, we present Deep Graph Kernels (DGK), a unified framework to learn latent representations of sub-structures for graphs, inspired by latest advancements in language modeling and deep learning.

Graph Classification Language Modelling

WordRank: Learning Word Embeddings via Robust Ranking

2 code implementations EMNLP 2016 Shihao Ji, Hyokun Yun, Pinar Yanardag, Shin Matsushima, S. V. N. Vishwanathan

Then, based on this insight, we propose a novel framework WordRank that efficiently estimates word representations via robust ranking, in which the attention mechanism and robustness to noise are readily achieved via the DCG-like ranking losses.

Learning Word Embeddings Word Similarity

The Structurally Smoothed Graphlet Kernel

no code implementations3 Mar 2014 Pinar Yanardag, S. V. N. Vishwanathan

This vector representation can be used in a variety of applications, such as, for computing similarity between graphs.

Cannot find the paper you are looking for? You can Submit a new open access paper.