SHSamet Hicsonmez

Samet Hicsonmez

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Samet Hicsonmez has a total of 8 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    NSYNC: Negative Synthetic Image Generation for Contrastive Training to Improve Stylized Text-To-Image Translation

    by Samet Hicsonmez, Serkan Ozturk, Pinar Duygulu

    Text-conditioned image generation models can synthesize highly realistic images, but they fail to capture specific styles. Fine-tuning on style-specific datasets alone is not always sufficient to capture the defining characteristics of a target style. In this work, we introduce NSYNC, a contrastive training framework for improving the stylization capability of large text-to-image diffusion models through synthetic negative samples.Unlike conventional synthetic data generation, which typically augments real datasets with additional positive examples, NSYNC constructs synthetic negative sets and trains them jointly with real positive images in a novel contrastive training scheme. Specifically, we compute gradients from positive and negative samples and refine the positive gradient by removing its projection onto the negative gradient. The resulting orthogonal update suppresses attributes shared by both sets and emphasizes features that are more distinctive of the target style.Extensive experiments across multiple painter and illustrator styles demonstrate that NSYNC consistently improves over baseline methods in both quantitative metrics and qualitative evaluations. Our code is available at \url{https://github.com/giddyyupp/NSYNC}.

  • WAIT: Feature warping for animation to illustration video translation using GANs

    by Samet Hicsonmez, Nermin Samet, Fidan Samet, Oguz Bakir, Emre Akbas, Pinar Duygulu


  • Open Access

    Improving Sketch Colorization using Adversarial Segmentation Consistency

    by Samet Hicsonmez, Nermin Samet, Emre Akbas, Pinar Duygulu

    We propose a new method for producing color images from sketches. Current solutions in sketch colorization either necessitate additional user instruction or are restricted to the "paired" translation strategy. We leverage semantic image segmentation from a general-purpose panoptic segmentation network to generate an additional adversarial loss function. The proposed loss function is compatible with any GAN model. Our method is not restricted to datasets with segmentation labels and can be applied to unpaired translation tasks as well. Using qualitative, and quantitative analysis, and based on a user study, we demonstrate the efficacy of our method on four distinct image datasets. On the FID metric, our model improves the baseline by up to 35 points. Our code, pretrained models, scripts to produce newly introduced datasets and corresponding sketch images are available at https://github.com/giddyyupp/AdvSegLoss.

  • HoughNet: Integrating Near and Long-Range Evidence for Visual Detection

    by Nermin Samet, Samet Hicsonmez, Emre Akbas

    This paper presents HoughNet, a one-stage, anchor-free, voting-based, bottom-up object detection method. Inspired by the Generalized Hough Transform, HoughNet determines the presence of an object at a certain location by the sum of the votes cast on that location. Votes are collected from both near and long-distance locations based on a log-polar vote field. Thanks to this voting mechanism, HoughNet is able to integrate both near and long-range, class-conditional evidence for visual recognition, thereby generalizing and enhancing current object detection methodology, which typically relies on only local evidence. On the COCO dataset, HoughNet's best model achieves 46.4 $AP$ (and 65.1 $AP_{50}$), performing on par with the state-of-the-art in bottom-up object detection and outperforming most major one-stage and two-stage methods. We further validate the effectiveness of our proposal in other visual detection tasks, namely, video object detection, instance segmentation, 3D object detection and keypoint detection for human pose estimation, and an additional “labels to photo” image generation task, where the integration of our voting module consistently improves performance in all cases. Code is available at https://github.com/nerminsamet/houghnet.


  • Open Access

    HoughNet: Integrating near and long-range evidence for visual detection

    by Nermin Samet, Samet Hicsonmez, Emre Akbas

    This paper presents HoughNet, a one-stage, anchor-free, voting-based, bottom-up object detection method. Inspired by the Generalized Hough Transform, HoughNet determines the presence of an object at a certain location by the sum of the votes cast on that location. Votes are collected from both near and long-distance locations based on a log-polar vote field. Thanks to this voting mechanism, HoughNet is able to integrate both near and long-range, class-conditional evidence for visual recognition, thereby generalizing and enhancing current object detection methodology, which typically relies on only local evidence. On the COCO dataset, HoughNet's best model achieves $46.4$ $AP$ (and $65.1$ $AP_{50}$), performing on par with the state-of-the-art in bottom-up object detection and outperforming most major one-stage and two-stage methods. We further validate the effectiveness of our proposal in other visual detection tasks, namely, video object detection, instance segmentation, 3D object detection and keypoint detection for human pose estimation, and an additional "labels to photo" image generation task, where the integration of our voting module consistently improves performance in all cases. Code is available at https://github.com/nerminsamet/houghnet.

  • Open Access

    Reducing Label Noise in Anchor-Free Object Detection

    by Nermin Samet, Samet Hicsonmez, Emre Akbas

    Current anchor-free object detectors label all the features that spatially fall inside a predefined central region of a ground-truth box as positive. This approach causes label noise during training, since some of these positively labeled features may be on the background or an occluder object, or they are simply not discriminative features. In this paper, we propose a new labeling strategy aimed to reduce the label noise in anchor-free detectors. We sum-pool predictions stemming from individual features into a single prediction. This allows the model to reduce the contributions of non-discriminatory features during training. We develop a new one-stage, anchor-free object detector, PPDet, to employ this labeling strategy during training and a similar prediction pooling method during inference. On the COCO dataset, PPDet achieves the best performance among anchor-free top-down detectors and performs on-par with the other state-of-the-art methods. It also outperforms all major one-stage and two-stage methods in small object detection (${AP}_{S}$ $31.4$). Code is available at https://github.com/nerminsamet/ppdet

  • Open Access

    GANILLA: Generative Adversarial Networks for Image to Illustration Translation

    by Samet Hicsonmez, Nermin Samet, Emre Akbas, Pinar Duygulu

    In this paper, we explore illustrations in children's books as a new domain in unpaired image-to-image translation. We show that although the current state-of-the-art image-to-image translation models successfully transfer either the style or the content, they fail to transfer both at the same time. We propose a new generator network to address this issue and show that the resulting network strikes a better balance between style and content. There are no well-defined or agreed-upon evaluation metrics for unpaired image-to-image translation. So far, the success of image translation models has been based on subjective, qualitative visual comparison on a limited number of images. To address this problem, we propose a new framework for the quantitative evaluation of image-to-illustration models, where both content and style are taken into account using separate classifiers. In this new evaluation framework, our proposed model performs better than the current state-of-the-art models on the illustrations dataset. Our code and pretrained models can be found at https://github.com/giddyyupp/ganilla.

  • Could we create a training set for image captioning using automatic translation?

    by Nermin Samet, Samet Hicsonmez, Pinar Duygulu, Emre Akbas

    Automatic image captioning has received increasing attention in recent years. Although there are many English datasets developed for this problem, there is only one Turkish dataset and it is very small compared to its English counterparts. Creating a new dataset for image captioning is a very costly and time consuming task. This work is a first step towards transferring the available, large English datasets into Turkish. We translated English captioning datasets into Turkish by using an automated translation tool and we trained an image captioning model on the automatically obtained Turkish captions. Our experiments show that this model yields the best performance so far on Turkish captioning.