Search

In order to solve the problem that the existing 3D model recognition methods based on deep learning lack the contextual fine-grained local features of 3D models, which may cause the confusion of recognition with very similar geometric shapes and slightly different local details, a 3D model recognition method based on deep graph attention convolutional neural network is proposed. Firstly, the fine-grained local features of 3D models are mined by introducing a neighborhood selection mechanism. Secondly, the multi-scale spatial context information is captured by a spatial context coding mechanism, and is compensated with fine-grained local features to enhance the completeness of features. Finally, a multi-head mechanism is adopted to make the graph attention convolution layer aggregate features of multiple single-head to enhance the richness of features. In addition, a selective dropout algorithm is designed to prevent network overfitting, which ranks the importance of neurons according to value of the measure-ment weight, and intelligently discards those with lower importance. The accuracy of 3D model recognition on the ModelNet40 dataset of the algorithm in this paper reaches 92.6%, and the network complexity is low. The trade-off between accuracy rate of 3D model recognition and network complexity achieved by the proposed algorithm is superior to contemporary mainstream methods.

Current video object segmentation approaches primarily rely on frame-wise appearance information to perform matching. Despite significant progress, reliable matching becomes challenging due to rapid changes of the object's appearance over time. Moreover, previous matching mechanisms suffer from redundant computation and noise interference as the number of accumulated frames increases. In this paper, we introduce a multi-frame spatio-temporal context memory (STCM) network to exploit discriminative spatio-temporal cues in multiple adjacent frames by utilizing a multi-frame context interaction module (MCI) for memory construction. Based on the proposed MCI module, a sparse group memory reader is developed to enable efficient sparse matching during memory reading. Our proposed method is generic and achieves state-of-the-art performance with real-time speed on benchmark datasets such as DAVIS and YouTube-VOS. In addition, our model exhibits robustness to sparse videos with low frame rates.

Recently, memory-based methods have achieved remarkable progress in video object segmentation. However, the segmentation performance is still limited by error accumulation and redundant memory, primarily because of 1) the semantic gap caused by similarity matching and memory reading via heterogeneous key-value encoding; 2) the continuously growing and inaccurate memory through directly storing unreliable predictions of all previous frames. To address these issues, we propose an efficient, effective, and robust segmentation method based on Isogenous Memory Sampling and Frame-Relation mining (IMSFR). Specifically, by utilizing an isogenous memory sampling module, IMSFR consistently conducts memory matching and reading between sampled historical frames and the current frame in an isogenous space, minimizing the semantic gap while speeding up the model through an efficient random sampling. Furthermore, to avoid key information loss during the sampling process, we further design a frame-relation temporal memory module to mine inter-frame relations, thereby effectively preserving contextual information from the video sequence and alleviating error accumulation. Extensive experiments demonstrate the effectiveness and efficiency of the proposed IMSFR method. In particular, our IMSFR achieves state-of-the-art performance on six commonly used benchmarks in terms of region similarity & contour accuracy and speed. Our model also exhibits strong robustness against frame sampling due to its large receptive field.