Coarse Mask Guided Interactive Object Segmentation.

Interactive object segmentation aims to produce object masks with user interactions, such as clicks, bounding boxes, and scribbles. Click point is the most popular interactive cue for its efficiency, and related deep learning methods have attracted lots of interest in recent years. Most works encode click points as gaussian maps and concatenate them with images as the model's input. However, the spatial and semantic information of gaussian maps would be noised through multiple convolution layers and won't be fully exploited by top layers for mask prediction. To pass click information to top layers exactly and efficiently, we propose a coarse mask guided model (CMG) which predicts coarse masks with a coarse module to guide the object mask prediction. Specifically, the coarse module encodes user clicks as query features and enriches their semantic information with backbone features through transformer layers, coarse masks are generated based on the enriched query feature and fed into CMG's decoder. Benefiting from the efficiency of transformer, CMG's coarse module and decoder module are lightweight and computationally efficient, making the interaction process more smooth. Experiments on several segmentation benchmarks demonstrate the effectiveness of our method, and we get new state-of-the-art results compared with previous works.

MMT: Cross Domain Few-Shot Learning via Meta-Memory Transfer.

Few-shot learning aims to recognize novel categories solely relying on a few labeled samples, with existing few-shot methods primarily focusing on the categories sampled from the same distribution. Nevertheless, this assumption cannot always be ensured, and the actual domain shift problem significantly reduces the performance of few-shot learning. To remedy this problem, we investigate an interesting and challenging cross-domain few-shot learning task, where the training and testing tasks employ different domains. Specifically, we propose a Meta-Memory scheme to bridge the domain gap between source and target domains, leveraging style-memory and content-memory components. The former stores intra-domain style information from source domain instances and provides a richer feature distribution. The latter stores semantic information through exploration of knowledge of different categories. Under the contrastive learning strategy, our model effectively alleviates the cross-domain problem in few-shot learning. Extensive experiments demonstrate that our proposed method achieves state-of-the-art performance on cross-domain few-shot semantic segmentation tasks on the COCO-20 i, PASCAL-5 i, FSS-1000, and SUIM datasets and positively affects few-shot classification tasks on Meta-Dataset.

Memory-Based Cross-Image Contexts for Weakly Supervised Semantic Segmentation.

Weakly supervised semantic segmentation (WSSS) trains segmentation models by only weak labels, aiming to save the burden of expensive pixel-level annotations. This paper tackles the WSSS problem of utilizing image-level labels as the weak supervision. Previous approaches address this problem by focusing on generating better pseudo-masks from weak labels to train the segmentation model. However, they generally only consider every single image and overlook the potential cross-image contexts. We emphasize that the cross-image contexts among a group of images can provide complementary information for each other to obtain better pseudo-masks. To effectively employ cross-image contexts, we develop an end-to-end cross-image context module containing a memory bank mechanism and a transformer-based cross-image attention module. The former extracts cross-image contexts online from the feature encodings of input images and stores them as the memory. The latter mines useful information from the memorized contexts to provide the original queries with additional information for better pseudo-mask generation. We conduct detailed experiments on the Pascal VOC 2012 and the COCO dataset to demonstrate the advantage of utilizing cross-image contexts. Besides, state-of-the-art performance is also achieved. Codes are available at https://github.com/js-fan/MCIC.git.

Autistic traits heighten sensitivity to rejection-induced social pain.

Autistic traits-subclinical forms of characteristics associated with autism spectrum disorders-are associated with poor social interactions and high risks for mental health disorders. We hypothesized that altered sensitivity to social rejection is an important contributor to psychological distress observed among individuals with high autistic traits. Experiment 1 adopted a social-judgment task and compared behavioral and neural activity in response to social rejection between participants exhibiting either high or low autistic traits (HAT and LAT, respectively). Rejection-induced hurt feelings, P3 amplitudes, and θ-oscillation magnitudes were greater in the HAT group than in the LAT group. Mediation analysis indicated that autistic traits heighten rejection-induced social pain through increasing frontal-midline θ-oscillations. Responses to nonsocial feedback in the age-judgment task were comparable, confirming that the between-group differences were specific to social negative feedback. Experiment 2 assessed the association between autistic traits, rejection sensitivity, and psychological distress among randomly recruited participants. Results showed that autistic traits affected depressive/anxious symptomatology partially through heightened rejection sensitivity. Therefore, autistic traits heighten sensitivity to rejection-induced social pain that leads to psychological distress. This finding will help facilitate the development of strategies for coping with social pain and improving mental health for individuals with high autistic traits.

Single-dose testosterone administration modulates instant empathic responses to others' pain: An EEG study.

Whether or not testosterone can impair empathy remains unclear in the literature. Given that empathic responses to others' emotional experiences depend strongly upon top-down controlled mechanisms of attention, here we investigated whether the effects of testosterone administration on pain empathy could be modulated by manipulating attention. We used a double-blind, placebo-controlled within-participant design, in which either testosterone or placebo was administrated in separate sessions. Images depicting painful or nonpainful scenes were presented to induce instant empathic responses. Experiment 1 adopted the pain-judgment and hands-counting tasks to direct attention toward painful or nonpainful aspect of the images, respectively. Experiment 2 employed the pain-rating task to estimate affective and cognitive aspects of pain empathy. When discriminating nonpainful aspects of the images in the hands-counting task, accuracies were lower and empathic late positive potential responses were greater in testosterone sessions than in placebo sessions. This suggested that testosterone enhanced empathic responses to task-irrelevant pain-related features, which interfered with task performance. When providing empathic ratings to the images in the pain-rating task, empathic event-related potentials in the early stage were only observed in the testosterone session. This suggested that testosterone facilitated automatic affective reactivity to others' pain when elaborately processing empathic stimuli. Nevertheless, when discriminating painful aspects of the images in the pain-judgment task, we did not observe any significant differences between the two sessions. These results demonstrated that testosterone effects on enhancing brain reactivity to empathic stimuli were dependent upon task demands deploying attention allocation. The enhancement likely arose from the altered brain state (e.g., increased vigilance and arousal levels) after testosterone administration, as evidenced by the reduced amplitude of spontaneous α-oscillation recorded before the onset of the images. It expands our understanding of the neurobiological mechanisms that affect empathy, and highlights the role of testosterone.