Search Results for author:
Found 25 papers, 13 papers with code
BAM-DETR: Boundary-Aligned Moment Detection Transformer for Temporal Sentence Grounding in Videos
However, they suffer from the issue of center misalignment raised by the inherent ambiguity of moment centers, leading to inaccurate predictions.
Ranked #1 on
Natural Language Moment Retrieval
on TACoS
Small Objects Matters in Weakly-supervised Semantic Segmentation
Weakly-supervised semantic segmentation (WSSS) performs pixel-wise classification given only image-level labels for training.
Weakly supervised Semantic Segmentation
Weakly-Supervised Semantic Segmentation
Improving Diversity in Zero-Shot GAN Adaptation with Semantic Variations
Due to the data absence, the textual description of the target domain and the vision-language models, e. g., CLIP, are utilized to effectively guide the generator.
AesPA-Net: Aesthetic Pattern-Aware Style Transfer Networks
To deliver the artistic expression of the target style, recent studies exploit the attention mechanism owing to its ability to map the local patches of the style image to the corresponding patches of the content image.
Decomposed Cross-modal Distillation for RGB-based Temporal Action Detection
Temporal action detection aims to predict the time intervals and the classes of action instances in the video.
BallGAN: 3D-aware Image Synthesis with a Spherical Background
To solve this problem, we propose to approximate the background as a spherical surface and represent a scene as a union of the foreground placed in the sphere and the thin spherical background.
Source-free Subject Adaptation for EEG-based Visual Recognition
In this paper, we introduce a novel and practical problem setup, namely source-free subject adaptation, where the source subject data are unavailable and only the pre-trained model parameters are provided for subject adaptation.
Exploiting Shape Cues for Weakly Supervised Semantic Segmentation
To this end, previous methods adopt the common pipeline: they generate pseudo masks from class activation maps (CAMs) and use such masks to supervise segmentation networks.
Exploiting Domain Transferability for Collaborative Inter-level Domain Adaptive Object Detection
Domain adaptation for object detection (DAOD) has recently drawn much attention owing to its capability of detecting target objects without any annotations.
Fair Contrastive Learning for Facial Attribute Classification
Through extensive experiments on CelebA and UTK Face, we validate that the proposed method significantly outperforms SupCon and existing state-of-the-art methods in terms of the trade-off between top-1 accuracy and fairness.
Inter-subject Contrastive Learning for Subject Adaptive EEG-based Visual Recognition
This paper tackles the problem of subject adaptive EEG-based visual recognition.
Subject Adaptive EEG-based Visual Recognition
It limits recognition systems to work only for the subjects involved in model training, which is undesirable for real-world scenarios where new subjects are frequently added.
Feature Stylization and Domain-aware Contrastive Learning for Domain Generalization
To these ends, we propose a novel domain generalization framework where feature statistics are utilized for stylizing original features to ones with novel domain properties.
Ranked #34 on
Domain Generalization
on Office-Home
Learning Action Completeness from Points for Weakly-supervised Temporal Action Localization
To learn completeness from the obtained sequence, we introduce two novel losses that contrast action instances with background ones in terms of action score and feature similarity, respectively.
Ranked #1 on
Weakly Supervised Action Localization
on THUMOS’14
Weakly Supervised Action Localization
Weakly-supervised Temporal Action Localization
+1
Domain-Aware Universal Style Transfer
To this end, we design a novel domainness indicator that captures the domainness value from the texture and structural features of reference images.
Continuous Face Aging Generative Adversarial Networks
Face aging is the task aiming to translate the faces in input images to designated ages.
ArrowGAN : Learning to Generate Videos by Learning Arrow of Time
Training GANs on videos is even more sophisticated than on images because videos have a distinguished dimension: time.
Contrastive Attention Maps for Self-Supervised Co-Localization
The goal of unsupervised co-localization is to locate the object in a scene under the assumptions that 1) the dataset consists of only one superclass, e. g., birds, and 2) there are no human-annotated labels in the dataset.
In-sample Contrastive Learning and Consistent Attention for Weakly Supervised Object Localization
Furthermore, we propose foreground consistency loss that penalizes earlier layers producing noisy attention regarding the later layer as a reference to provide them with a sense of backgroundness.
Weakly-supervised Temporal Action Localization by Uncertainty Modeling
Experimental results show that our uncertainty modeling is effective at alleviating the interference of background frames and brings a large performance gain without bells and whistles.
Learning Texture Invariant Representation for Domain Adaptation of Semantic Segmentation
However, due to the domain gap between synthetic domain and real domain, it is challenging for a model trained with synthetic data to generalize to real data.
Background Suppression Network for Weakly-supervised Temporal Action Localization
This formulation does not fully model the problem in that background frames are forced to be misclassified as action classes to predict video-level labels accurately.
Ranked #9 on
Weakly Supervised Action Localization
on ActivityNet-1.2
(mAP@0.5 metric)
Weakly Supervised Action Localization
Weakly-supervised Temporal Action Localization
+1
Contextual Action Cues from Camera Sensor for Multi-Stream Action Recognition
However, depending on the action characteristics, contextual information, such as the existence of specific objects or globally-shared information in the image, becomes vital information to define the action.
Ranked #17 on
Action Recognition
on HMDB-51
