Search Results for author:
Found 29 papers, 12 papers with code
Would Deep Generative Models Amplify Bias in Future Models?
We investigate the impact of deep generative models on potential social biases in upcoming computer vision models.
LayoutFlow: Flow Matching for Layout Generation
Finding a suitable layout represents a crucial task for diverse applications in graphic design.
Multimodal Color Recommendation in Vector Graphic Documents
Additionally, it is applicable for another color recommendation task, full palette generation, which generates a complete color palette corresponding to the given text.
Toward Verifiable and Reproducible Human Evaluation for Text-to-Image Generation
Human evaluation is critical for validating the performance of text-to-image generative models, as this highly cognitive process requires deep comprehension of text and images.
Towards Flexible Multi-modal Document Models
Creative workflows for generating graphical documents involve complex inter-related tasks, such as aligning elements, choosing appropriate fonts, or employing aesthetically harmonious colors.
LayoutDM: Discrete Diffusion Model for Controllable Layout Generation
Controllable layout generation aims at synthesizing plausible arrangement of element bounding boxes with optional constraints, such as type or position of a specific element.
Generative Colorization of Structured Mobile Web Pages
The web page colorization problem is then formalized as a task of estimating plausible color styles for a given web page content with a given hierarchical structure of the elements.
Contrastive Losses Are Natural Criteria for Unsupervised Video Summarization
Video summarization aims to select the most informative subset of frames in a video to facilitate efficient video browsing.
Video Summarization Overview
With the broad growth of video capturing devices and applications on the web, it is more demanding to provide desired video content for users efficiently.
Color Recommendation for Vector Graphic Documents based on Multi-Palette Representation
We train the model and build a color recommendation system on a large-scale dataset of vector graphic documents.
Learning More May Not Be Better: Knowledge Transferability in Vision and Language Tasks
Is more data always better to train vision-and-language models?
Does Robustness on ImageNet Transfer to Downstream Tasks?
For object detection and semantic segmentation, we find that a vanilla Swin Transformer, a variant of Vision Transformer tailored for dense prediction tasks, transfers robustness better than Convolutional Neural Networks that are trained to be robust to the corrupted version of ImageNet.
AxIoU: An Axiomatically Justified Measure for Video Moment Retrieval
First, it is rank-insensitive: It ignores the rank positions of successfully localised moments in the top-$K$ ranked list by treating the list as a set.
Optimal Correction Cost for Object Detection Evaluation
OC-cost computes the cost of correcting detections to ground truths as a measure of accuracy.
Transferring Domain-Agnostic Knowledge in Video Question Answering
Video question answering (VideoQA) is designed to answer a given question based on a relevant video clip.
Constrained Graphic Layout Generation via Latent Optimization
We optimize using the latent space of an off-the-shelf layout generation model, allowing our approach to be complementary to and used with existing layout generation models.
Attending Self-Attention: A Case Study of Visually Grounded Supervision in Vision-and-Language Transformers
The impressive performances of pre-trained visually grounded language models have motivated a growing body of research investigating what has been learned during the pre-training.
A Picture May Be Worth a Hundred Words for Visual Question Answering
This paper delves into the effectiveness of textual representations for image understanding in the specific context of VQA.
Scalable Personalised Item Ranking through Parametric Density Estimation
However, such methods have two main drawbacks particularly in large-scale applications; (1) the pairwise approach is severely inefficient due to the quadratic computational cost; and (2) even recent model-based samplers (e. g. IRGAN) cannot achieve practical efficiency due to the training of an extra model.
Density-Ratio Based Personalised Ranking from Implicit Feedback
Learning from implicit user feedback is challenging as we can only observe positive samples but never access negative ones.
Alleviating Cold-Start Problems in Recommendation through Pseudo-Labelling over Knowledge Graph
Solving cold-start problems is indispensable to provide meaningful recommendation results for new users and items.
Uncovering Hidden Challenges in Query-Based Video Moment Retrieval
In this paper, we present a series of experiments assessing how well the benchmark results reflect the true progress in solving the moment retrieval task.
A Dataset and Baselines for Visual Question Answering on Art
Our dataset inherently consists of visual (painting-based) and knowledge (comment-based) questions.
Knowledge-Based Visual Question Answering in Videos
We propose a novel video understanding task by fusing knowledge-based and video question answering.
KnowIT VQA: Answering Knowledge-Based Questions about Videos
We propose a novel video understanding task by fusing knowledge-based and video question answering.
Rethinking the Evaluation of Video Summaries
Video summarization is a technique to create a short skim of the original video while preserving the main stories/content.
iParaphrasing: Extracting Visually Grounded Paraphrases via an Image
These extracted VGPs have the potential to improve language and image multimodal tasks such as visual question answering and image captioning.
Video Summarization using Deep Semantic Features
For this, we design a deep neural network that maps videos as well as descriptions to a common semantic space and jointly trained it with associated pairs of videos and descriptions.
Learning Joint Representations of Videos and Sentences with Web Image Search
In description generation, the performance level is comparable to the current state-of-the-art, although our embeddings were trained for the retrieval tasks.
