Danmaku, a system of scene-aligned, time-synced, floating comments, can augment video content to create g'collective knowledge'. However, its chaotic nature often hinders viewers from effectively assimilating the collective knowledge, especially in knowledge-intensive science videos. With a formative study, we examined viewers' practices for processing collective knowledge and the specific barriers they encountered. Building on these insights, we designed a processing pipeline to filter, classify, and cluster danmaku, leading to the development of CoKnowledge - a tool incorporating a video abstract, knowledge graphs, and supplementary danmaku features to support viewers' assimilation of collective knowledge in science videos. A within-subject study (N=24) showed that CoKnowledge significantly enhanced participants' comprehension and recall of collective knowledge compared to a baseline with unprocessed live comments. Based on our analysis of user interaction patterns and feedback on design features, we presented design considerations for developing similar support tools.

Cell instance segmentation in cytology images has significant importance for biology analysis and cancer screening, while remains challenging due to 1) the extensive over-lapping translucent cell clusters that cause the ambigu-ous boundaries, and 2) the confusion of mimics and de-bris as nuclei. In this work, we proposed a De-overlapping Network (DoNet) in a decompose-and-recombined strategy. A Dual-path Region Segmentation Module (DRM) explicitly decomposes the cell clusters into intersection and complement regions, followed by a Semantic Consistency-guided Recombination Module (CRM) for integration. To further introduce the containment relationship of the nu-cleus in the cytoplasm, we design a Mask-guided Region Proposal Strategy (MRP) that integrates the cell attention maps for inner-cell instance prediction. We validate the proposed approach on ISBI2014 and CPS datasets. Ex-periments show that our proposed DoNet significantly outperforms other state-of-the-art (SOTA) cell instance segmentation methods. The code is available at https://github.com/DeepDoNet/DoNet.