Cross-modal retrieval aims to match videos and texts by mapping them into a shared feature space. Most existing approaches achieve alignment through contrastive learning based on one-to-one supervised pairs. However, these methods rely too much on supervised signals and do not fully use the unsupervised semantic relationships within each modality. As a result, samples that are semantically similar may be spread in the shared space, which hurts retrieval performance. To solve this problem, we propose a method called Leveraging Intra-Modal Consistency for Cross-Modal Alignment and Retrieval (LICA). Our method introduces a consistency constraint between intra-modal similarities and cross-modal similarity distributions. In this way, samples that are close in meaning stay closer together in the shared space. Experiments on standard text-video retrieval benchmarks show that LICA helps optimize the distribution of the cross-modal feature space and improves retrieval accuracy.

The paper presents a spatial-color model of object and develops an efficient visual tracking algorithm based on particle filter. This spatial-color model captures richer information than the general color histogram because it incorporates spatial distribution of pixels in addition to color. In order to fast compute the weight of each particle, the Integral Images for computation of histogram, mean and variance are proposed, with which the parameters of spatial-color histograms can be computed by using only simple arithmetic operations and thus the computation load is reduce to linear complexity. Experiments on real image sequences show our method results in more accurate tracking than a standard color particle filter.