16 accepted papers
Multi-task learning often suffers from gradient conflicts, leading to unfair optimization and degraded overall performance. To address this, we present SVFair, a Shapley value-based framework for fair gradient aggregation. We propose two scalable geometric conflict metrics: VolDet, a gram determinan…
Grasp and in-hand manipulation of dexterous hands are essential for daily operational tasks. In this study, an optimized design method for a novel anthropomorphic dexterous hand is proposed, which improves the comprehensive performance of grasp and manipulability of the dexterous hand. The constrain
This paper presents the first study on adapting the visual in-context learning (V-ICL) paradigm to optical character recognition tasks, specifically focusing on text removal and segmentation. Most existing V-ICL generalists employ a reasoning-as-reconstruction approach: they turn to using a straight…
The automatic extinguishing strategy (AES) is the core of the decision-making system for intelligent firefighting robots. Inspired by the fire extinguishing action of firefighters, designing a vision-based end-to-end AES aligns with human intuition. However, the cost of training agents to learn AES
Existing video watermarking embeds robust watermarks in each frame of the video for copyright protection and tracking. However, just as any content written on a blank paper is easily perceived, embedding watermarks in the texture-poor frames impairs imperceptibility. Common geometric attacks such as…
Audio-visual segmentation (AVS) aims to segment the sounding objects in video frames. Although great progress has been witnessed we experimentally reveal that current methods reach marginal performance gain within the use of the unlabeled frames leading to the underutilization issue. To fully explor…
With the rapid development of mobile Internet and video applications, robust video watermarking technology has become a focal area of research for protecting and tracking intellectual property rights in digital media. An important characteristic of video is that it has both spatial and temporal prop…
Multimodal models trained on modality-complete data are plagued with severe performance degradation when encountering modality-missing data. Prevalent cross-modal knowledge distillation-based methods precisely align the representation of modality-missing data and that of its modality-complete counte…
Open-vocabulary semantic segmentation is a challenging task that requires segmenting novel object categories at inference time. Recent works explore vision-language pre-training to handle this task, but suffer from unrealistic assumptions in practical scenarios, i.e., low-quality textual category n…
Five-hundred-meter aperture spherical radio telescope (FAST) is a very famous large-scale scientific facility with excellent performance for astronomical observation in the world, but it currently fails to observe the center of the Milky Way Galaxy due to the limited observation angle that is affect…
Single-frame infrared small target (SIRST) detection aims at separating small targets from clutter backgrounds on infrared images. Recently, deep learning based methods have achieved promising performance on SIRST detection, but at the cost of a large amount of training data with expensive pixel-lev…
This paper studies the problem of weakly open-vocabulary semantic segmentation (WOVSS), which learns to segment objects of arbitrary classes using mere image-text pairs. Existing works turn to enhance the vanilla vision transformer by introducing explicit grouping recognition, i.e., employing severa…
The lightweight design of robots is an important factor in the field of human-robot interaction. Thus, a lightweight 7-DOF cable-driven manipulator is proposed and manufactured, including a shoulder joint with 3-DOF, an elbow joint with 1-DOF, and a wrist joint with 3-DOF. The offset design of the s
Partial-label learning (PLL) solves the multi-class classification problem, where each training instance is assigned a set of candidate labels that include the true label. Recent advances showed that PLL can be compatible with deep neural networks, which achieved state-of-the-art performance. Howeve…
Weakly Supervised Semantic Segmentation (WSSS) based on image-level labels has been greatly advanced by exploiting the outputs of Class Activation Map (CAM) to generate the pseudo labels for semantic segmentation. However, CAM merely discovers seeds from a small number of regions, which may be insuf…
Generalized zero-shot learning is a significant topic but faced with bias problem, which leads to unseen classes being easily misclassified into seen classes. Hence we propose a embedding model called co-representation network to learn a more uniform visual embedding space that effectively alleviate…