BEV-CLIP: Multi-modal BEV Retrieval Methodology for Complex Scene in Autonomous Driving

The demand for the retrieval of complex scene data in autonomous driving is increasing, especially as passenger vehicles have been equipped with the ability to navigate urban settings, with the imperative to address long-tail scenarios. Meanwhile, under the pre-existing two dimensional image retrieval method, some problems may arise with scene retrieval, such as lack of global feature representation and subpar text retrieval ability. To address these issues, we have proposed \textbf{BEV-CLIP}, the first multimodal Bird's-Eye View(BEV) retrieval methodology that utilizes descriptive text as an input to retrieve corresponding scenes. This methodology applies the semantic feature extraction abilities of a large language model (LLM) to facilitate zero-shot retrieval of extensive text descriptions, and incorporates semi-structured information from a knowledge graph to improve the semantic richness and variety of the language embedding. Our experiments result in 87.66% accuracy on NuScenes dataset in text-to-BEV feature retrieval. The demonstrated cases in our paper support that our retrieval method is also indicated to be effective in identifying certain long-tail corner scenes.

Advertising strategy for profit-maximization: a novel practice on Tmall's online ads manager platforms

Ads manager platform gains popularity among numerous e-commercial vendors/advertisers. It helps advertisers to facilitate the process of displaying their ads to target customers. One of the main challenges faced by advertisers, especially small and medium-sized enterprises, is to configure their advertising strategy properly. An ineffective advertising strategy will bring too many ``just looking'' clicks and, eventually, generate high advertising expenditure unproportionally to the growth of sales. In this paper, we present a novel profit-maximization model for online advertising optimization. The optimization problem is constructed to find optimal set of features to maximize the probability that target customers buy advertising products. We further reformulate the optimization problem to a knapsack problem with changeable parameters, and introduce a self-adjusted algorithm for finding the solution to the problem. Numerical experiment based on statistical data from Tmall show that our proposed method can optimize the advertising strategy given expenditure budget effectively.