To enable autonomous mobile robots to navigate safely and efficiently in crowded environments, an effective way is to predict future occupancy states of their local surroundings to navigate proactively. However, most prior works focus on occupancy grid map (OGM) prediction for automous driving scenarios and typically generates deterministic single-mode predictions. To address these limitations, we present a diffusion model-based framework for stochastic and uncertainty-aware OGM prediction. Prior OGM predictors in crowded environment settings are limited to binary occupancy representations (occupied or free), which constrain their expressive power and ignore even basic semantic distinctions such as humans, static obstacles, and free space. Our framework extends traditional OGM predictions to include these fundamental semantic categories, enhancing environmental prediction, and facilitating downstream tasks. Additionally, we develop a Unity-based 3D pedestrian-robot simulator that supports teleoperation-enabled acquisition of automatically annotated autonomous mobile robot data. Experimental evaluations demonstrate that our framework achieves higher prediction accuracy on binary OGM prediction tasks across three publicly available datasets, while achieving promise results in semantic OGM prediction tasks using datasets collected from our simulation environments. Implementation and simulator are open source at https://github.com/daleigehhh/DiffOGMP.

With the rise of "Metaverse" and "Web 3.0", Non-Fungible Token (NFT) has emerged as a kind of pivotal digital asset, garnering significant attention. By the end of March 2024, more than 1.7 billion NFTs have been minted across various blockchain platforms. To effectively locate a desired NFT, conducting searches within a vast array of NFTs is essential. The challenge in NFT retrieval is heightened due to the high degree of similarity among different NFTs, regarding regional and semantic aspects. In this paper, we will introduce a benchmark dataset named "NFT Top1000 Visual-Text Dataset" (NFT1000), containing 7.56 million image-text pairs, and being collected from 1000 most famous PFP1 NFT collections2 by sales volume on the Ethereum blockchain. Based on this dataset and leveraging the CLIP series of pre-trained models as our foundation, we propose the dynamic masking fine-tuning scheme. This innovative approach results in a 7.4\% improvement in the top1 accuracy rate, while utilizing merely 13\% of the total training data (0.79 million vs. 6.1 million). We also propose a robust metric Comprehensive Variance Index (CVI) to assess the similarity and retrieval difficulty of visual-text pairs data. The dataset will be released as an open-source resource. For more details, please refer to: https://github.com/ShuxunoO/NFT-Net.git.