This makes 360 scene understanding tasks, e.g., segmentation and tracking, crucial for appications, such as autonomous driving, robotics. With the recent emergence of foundation models, the community is, however, impeded by the lack of large-scale, labelled real-world datasets. This is caused by the inherent spherical properties, e.g., severe distortion in polar regions, and content discontinuities, rendering the annotation costly yet complex. This paper introduces Leader360V, the first large-scale (10K+), labeled real-world 360 video datasets for instance segmentation and tracking. Our datasets enjoy high scene diversity, ranging from indoor and urban settings to natural and dynamic outdoor scenes.

Specifically, through clustering, we first identify 1,291 user-focused topics from the million-scale real text-to-video prompt dataset, VidProM. Then, we use these topics to retrieve videos from YouTube, split the retrieved videos into clips, the clips and with generate specified both brief topics, and we detailed are left captions with about for each 1.09 clip. million After video verifying clips. Our experiments reveal that (1) current 16 text-to-video models do not achieve consistent performance across all user-focused topics; and (2) a simple model trained on VideoUFO outperforms others on worst-performing topics. The dataset and code are publicly available here and here under the CCBY 4.0 License.

VLMEvaluation To evaluate two VLMs (Frozen in Time [1] and VideoCLIP [13]), we use a hybrid approach that leverages both prototypical networks [11] and the video-language similarity metrics learned by both models. Below, we show an ablation study where we use only the video prototype networks. We show the performance of using only language similarity in the few-shot case to demonstrate the effects of sample removal, and we also show the effects of our hybrid weighting scheme, where we weight the language embeddings five times more than the video embeddings when constructing the hybrid prototype (as opposed to equal weighting during the regular hybrid approach). Although we perform our ablation study with Frozen-in-Time, and use the same weighting scheme and prototype strategy for VideoCLIP as well. For this study, we show activity and sub-activity classification accuracy in the 5-shot case. We visualize whether a given method uses language, video, or both to create its prototype embeddings.

Tokenizer, serving as a translator to map the intricate visual data into a compact latent space, lies at the core of visual generative models.