Human-centric Emotional Video Captioning (H-EVC) aims to generate fine-grained, emotion-related sentences for human-based videos, enhancing the understanding of human emotions and facilitating human-computer emotional interaction. However, existing video captioning methods primarily focus on overall event content, often overlooking sufficient subtle emotional clues and interactions in videos. As a result, the generated captions frequently lack emotional information. To address this, we propose a novel Emotion-oriented Cross-modal Prompting and Alignment (ECPA) approach for large foundation models to enhance H-EVC accuracy by effectively modeling fine-grained visual-textual emotion clues and interactions. Using large foundation models, our ECPA introduces two learnable prompting strategies: visual emotion prompting (VEP) and textual emotion prompting (TEP), as well as an emotion-oriented cross-modal alignment (ECA) module. In VEP, we develop two-level learnable visual prompts, i.e., emotion recognition (ER)-level and action unit (AU)-level prompting, to assist pre-trained vision-language foundation models to attend to both coarse and fine emotion-related visual information in videos. In TEP, we correspondingly devise two-level learnable textual prompts, i.e., sentence-level emotional tokens, and word-level masked tokens, for obtaining both whole and local textual prompt representations related to emotions. To further facilitate the interaction and alignment of visual-textual emotion prompt representations, our ECA introduces another two levels of emotion-oriented prompt alignment learning mechanisms: the ER-sentence level and the AU-word level alignment losses. Both enhance the model's ability to capture and integrate both global and local cross-modal emotion semantics, thereby enabling the generation of fine-grained emotional linguistic descriptions in video captioning. Extensive experiments not only demonstrate that our ECPA outperforms existing state-of-the-art approaches on various H-EVC datasets (a relative improvement of 8.48% on MAFW and 10.58% on EmVidCap) by a large margin, but also support zero-shot tasks on two video captioning datasets (MSVD and MSRVTT), underscoring its applicability and generalization capabilities. The project page (including the dataset and demo) can be found in https://github.com/virtuesvvy/ECPA_Emotional_Video_Captioning.

User-generated geo-tagged photos (UGPs) have emerged as a valuable tool for analyzing large-scale tourist place emotions with unprecedented detail. This process involves extracting and analyzing human emotions associated with specific locations. However, previous studies have been limited to analyzing individual faces in the UGPs. This approach falls short of representing the contextual scene characteristics, such as environmental elements and overall scene context, which may contain implicit emotional knowledge. To address this issue, we propose an innovative computational framework for global tourist place emotion analysis leveraging UGPs. Specifically, we first introduce a Multi-view Graph Fusion Network (M-GFN) to effectively recognize multi-view emotions from UGPs, considering crowd emotions and scene implicit sentiment. After that, we designed an attraction-specific emotion index (AEI) to quantitatively measure place emotions based on the identified multi-view emotions at various tourist attractions with place types. Complementing the AEI, we employ the emotion intensity index (EII) and Pearson correlation coefficient (PCC) to deepen the exploration of the association between attraction types and place emotions. The synergy of AEI, EII, and PCC allows comprehensive attraction-specific place emotion extraction, enhancing the overall quality of tourist place emotion analysis. Extensive experiments demonstrate that our framework enhances existing place emotion analysis methods, and the M-GFN outperforms state-of-the-art emotion recognition methods. Our framework can be adapted for various geo-emotion analysis tasks, like recognizing and regulating workplace emotions, underscoring the intrinsic link between emotions and geographic contexts.

Under the background of global warming, seasonal drought has become frequent and intensified in many parts of the world in recent years. Drought is one of the most widespread and severe natural disasters, and poses a serious threat to normal sugarcane growth and yield. However, a deep understanding of sugarcane responses to drought stress remains limited, especially at a large spatial scale. In this work, we used the traditional vegetation index (enhanced vegetation index, EVI) and newly downscaled satellite solar-induced chlorophyll fluorescence (SIF) to investigate the impacts of drought on sugarcane in a major sugarcane-planting region of China (Chongzuo City, Southwest China). The results showed that Chongzuo City experienced an extremely severe drought event during the critical growth periods of sugarcane from August to November 2009. During the early stage of the 2009 drought, sugarcane SIF exhibited a quick negative response with a reduction of approximately 2.5% from the multiyear mean in late August 2009, while EVI was not able to capture the drought stress until late September 2009. Compared with EVI, sugarcane SIF shows more pronounced responses to drought stress during the later stage of drought, especially after late September 2009. SIF anomalies can closely capture the spatial and temporal dynamics of drought stress on sugarcane during this drought event. We also found that sugarcane SIF can provide earlier and much more pronounced physiological responses (as indicated by fluorescence yield) than structural responses (as indicated by the fraction of photosynthetically active radiation) to drought stress. Our results suggest that the satellite SIF has a great potential for sugarcane drought monitoring in a timely manner at a large spatial scale. These results are important for developing early warning models for sugarcane drought monitoring, and provide reliable information for developing measures to relieve the negative impacts of drought on sugarcane yield and regional economics.

Contour lines, as the primary elements of fundamental geospatial data, have long been a research focus for similarity measurement. With the evolution of cartographic generalization, the representation of contour lines across varying scales must maintain the consistency of specific information. Typically, the rationality of the generated results is assessed based on their similarity values. However, current measurements for measuring contour similarity predominantly focus on geometric and topological aspects, and are often less concerned with the terrain-specific similarities that are intrinsic to contour lines. Contour line groups contain a wealth of topographic information, and the similarity of their terrain features reflects both the variations in relief and the intrinsic nature of landform development. In this study, we propose a novel metric for assessing the similarity of contour line groups by considering topographic features, aiming to evaluate the similarity of contour line groups from a holistic perspective. First, we analyze and define the geometric, topological, and topographic similarity calculation metrics for contour line groups. Next, we apply the Analytic Hierarchy Process using ten criteria, which are encompassed by these three similarity metrics. To validate the effectiveness of the proposed metric, we select hillock areas within Suide County, China, as a case study for examining the similarity of contour line groups. The results demonstrate that the proposed metric provides a more precise quantitative framework for delineating the subtle differences and similarities among multi-source and multi-scale contour line groups within the overall similarity. Moreover, the metric also establishes a foundation for the quantitative assessment of surface morphology and the classification of geomorphological types.