Recent advancements in diffusion-based text-to-image (T2I) models have enabled the generation of high-quality and photorealistic images from text. However, they often exhibit societal biases related to gender, race, and socioeconomic status, thereby potentially reinforcing harmful stereotypes and shaping public perception in unintended ways. While existing bias mitigation methods demonstrate effectiveness, they often encounter attribute entanglement, where adjustments to attributes relevant to the bias (i.e., target attributes) unintentionally alter attributes unassociated with the bias (i.e., non-target attributes), causing undesirable distribution shifts. To address this challenge, we introduce Entanglement-Free Attention (EFA), a method that accurately incorporates target attributes (e.g., White, Black, and Asian) while preserving non-target attributes (e.g., background) during bias mitigation. At inference time, EFA randomly samples a target attribute with equal probability and adjusts the cross-attention in selected layers to incorporate the sampled attribute, achieving a fair distribution of target attributes. Extensive experiments demonstrate that EFA outperforms existing methods in mitigating bias while preserving non-target attributes, thereby maintaining the original model's output distribution and generative capacity.

Recent research has unveiled the development of customizing large-scale text-to-image models. These models bind a unique subject desired by a user to a specific token, using the token to generate the subject in various contexts. However, models from previous studies also bind elements unrelated to the subject's identity, such as common backgrounds or poses in the reference images. This often leads to conflicts between the token and the context of text prompts during inference, causing the model to fail to generate both the subject and the prompted context. In this work, we approach this issue from a data scarcity perspective and propose to augment the number of reference images through a novel self-distillation framework. Our framework selects high-quality samples from images generated by a teacher model and uses them in student training. Our framework can be applied to any models that suffer from the conflicts, and we demonstrate that our framework most effectively resolves the issue through comprehensive evaluations.

In the image classification task, deep neural networks frequently rely on bias attributes that are spuriously cor-related with a target class in the presence of dataset bias, resulting in degraded performance when applied to data without bias attributes. The task of debiasing aims to compel classifiers to learn intrinsic attributes that inher-ently define a target class rather than focusing on bias at-tributes. While recent approaches mainly focus on empha-sizing the learning of data samples without bias attributes (i.e., bias-conflicting samples) compared to samples with bias attributes (i.e., bias-aligned samples), they fall short of directly guiding models where to focus for learning in-trinsic features. To address this limitation, this paper pro-poses a method that provides the model with explicit spa-tial guidance that indicates the region of intrinsic features. We first identify the intrinsic features by investigating the class-discerning common features between a bias-aligned (BA) sample and a bias-conflicting (BC) sample (i.e., bias-contrastive pair). Next, we enhance the intrinsic features in the BA sample that are relatively under-exploited for pre-diction compared to the BC sample. To construct the bias-contrastive pair without using bias information, we intro-duce a bias-negative score that distinguishes BC samples from BA samples employing a biased model. The experi-ments demonstrate that our method achieves state-of-the-art performance on synthetic and real-world datasets with various levels of bias severity.

In image classification, debiasing aims to train a classifier to be less susceptible to dataset bias, the strong correlation between peripheral attributes of data samples and a target class. For example, even if the frog class in the dataset mainly consists of frog images with a swamp background (i.e., bias aligned samples), a debiased classifier should be able to correctly classify a frog at a beach (i.e., bias conflicting samples). Recent debiasing approaches commonly use two components for debiasing, a biased model fB and a debiased model fD. fB is trained to focus on bias aligned samples (i.e., overfitted to the bias) while fD is mainly trained with bias conflicting samples by concentrating on samples which fB fails to learn, leading fD to be less susceptible to the dataset bias. While the state of the art debiasing techniques have aimed to better train fD, we focus on training fB, an overlooked component until now. Our empirical analysis reveals that removing the bias conflicting samples from the training set for fB is important for improving the debiasing performance of fD. This is due to the fact that the bias conflicting samples work as noisy samples for amplifying the bias for fB since those samples do not include the bias attribute. To this end, we propose a simple yet effective data sample selection method which removes the bias conflicting samples to construct a bias amplified dataset for training fB. Our data sample selection method can be directly applied to existing reweighting based debiasing approaches, obtaining consistent performance boost and achieving the state of the art performance on both synthetic and real-world datasets.

Machine learning-based algorithms using fully convolutional networks (FCNs) have been a promising option for medical image segmentation. However, such deep networks silently fail if input samples are drawn far from the training data distribution, thus causing critical problems in automatic data processing pipelines. To overcome such out-of-distribution (OoD) problems, we propose a novel OoD score formulation and its regularization strategy by applying an auxiliary add-on classifier to an intermediate layer of an FCN, where the auxiliary module is helfpul for analyzing the encoder output features by taking their class information into account. Our regularization strategy train the module along with the FCN via the principle of outlier exposure so that our model can be trained to distinguish OoD samples from normal ones without modifying the original network architecture. Our extensive experiment results demonstrate that the proposed approach can successfully conduct effective OoD detection without loss of segmentation performance. In addition, our module can provide reasonable explanation maps along with OoD scores, which can enable users to analyze the reliability of predictions.

Deep neural networks for automatic image colorization often suffer from the color-bleeding artifact, a problematic color spreading near the boundaries between adjacent objects. Such color-bleeding artifacts debase the reality of generated outputs, limiting the applicability of colorization models in practice. Although previous approaches have attempted to address this problem in an automatic manner, they tend to work only in limited cases where a high contrast of gray-scale values are given in an input image. Alternatively, leveraging user interactions would be a promising approach for solving this color-breeding artifacts. In this paper, we propose a novel edge-enhancing network for the regions of interest via simple user scribbles indicating where to enhance. In addition, our method requires a minimal amount of effort from users for their satisfactory enhancement. Experimental results demonstrate that our interactive edge-enhancing approach effectively improves the color-bleeding artifacts compared to the existing baselines across various datasets.

This paper describes a joint response to the Call for Proposals by Samsung, Huawei, GoPro, and HiSilicon on Video Compression with Capability beyond HEVC/H.265, jointly issued by ITU-T SG16 Q.6 (VCEG) and ISO/IEC JTC1/SC29/WG11 (MPEG). In the proposed codec, the coding framework supports hierarchical splitting with binary and ternary trees and flexible coding order representations. Additionally, novel compression tools on inter/intra prediction, in-loop filtering, and entropy coding have been proposed. The proposed compression scheme provides significantly higher compression capability than the state-of-the-art HEVC/H.265 standard for SDR (Standard Dynamic Range) category while maintaining complexity acceptable for emerging applications. When all the proposed algorithmic tools are used, the proposed video codec achieves approximately 40% bit-saving for the SDR cetegory on average compared to HEVC/H.265 anchor.

Conventional video quality assessment methods are either full-, reduced-, or no-reference methods that need to access decoded videos. Hence, to calculate quality of decoded video in video coding regarding an image/video quality metric, complete encoding and decoding have to executed, which is computationally expensive. To address this problem, we propose to estimate quality of decoded videos from the original video only (i.e., only-reference) using convolutional neural network, as if the original video is encoded using a range of quantization parameter. The proposed network is shallow and can be trained to estimate various video quality metrics. Furthermore, among potential rate control applications using the proposed network, we demonstrate achieving a targeted decoded-video quality by selecting a proper quantization parameter before actually encoding.

Widespread of virtual reality technologies across entertainment formats has created a diverse infrastructure of related technologies as head-mount displays, dome screens and virtual reality multi-camera platforms. As omnidirectional content is processing pipeline is completely different form conventional planar video and involves multiple conversion steps which affect quality in a different way. As a result of our research we propose objective quality estimation methodology and a set of tools to evaluate different projection methods and coding tools for omnidirectional video content.

Scalability in video coding is an effective functionality to serve various video contents at different levels of resolution and quality. Based on the High Efficiency Video Coding (HEVC) standard which achieves superior compression performance compared to the H.264/AVC, this paper proposes a new scalable coding method with a coding unit (CU) structure prediction technique for HEVC-based quality scalability. The CU structure in enhancement layer (EL) is differentially encoded using that of the basement layer (BL) as a predictor. The binary values describing the CU structure of BL are operated exclusive-OR (XOR) with those in EL at each CU depth and position. Compared to the simulcast coding, simulation without having any residual prediction technique verifies that the proposed method gains in bit-saving by 0.4% on average.