Image-based vehicle re-identification (ReID) has witnessed much progress in recent years thanks to the advances of deep neural networks. However, most of existing works struggled to extract robust and discriminative features from a single image at each feedforward to represent one vehicle instance. We argue that images taken from distinct viewpoints, e.g., front and back, have significantly different appearances and patterns for recognition. In order to ‘`memorize’' each vehicle, existing models often have to capture consistent ``ID codes" from totally different view images, which causes learning difficulties. Additionally, we claim that correspondences among views, i.e., various vehicle parts observed from the identical image and the same part observed from different viewpoints, contribute to instance-level feature learning as well. Motivated by these observations, we propose to extract comprehensive instance-specific representations of the same vehicle from multiple views through modelling part-wise correlations. To this end, we present an efficient transformer-based framework to exploit both inner- and inter-view correlations for vehicle ReID. In specific, we first adopt a deep encoder to condense a series of patch embeddings from each view image. Then our efficient transformer, consisting of a distillation token and a noise token in addition to a regular class token, is constructed to enforce all patch embeddings to interact with each other regardless of whether they are taken from identical or different views. For inference, one testing image together with its augmented counterparts (pseudo views) are regarded as multi-view inputs and fed into our framework to capture its representations. We conduct extensive experiments on widely used vehicle ReID benchmarks, and our approach achieves the state-of-the-art performance, showing the effectiveness of our method.

Overexpression of antiapoptotic members of the Bcl-2 family proteins, such as Bcl-x_L and Mfl-1, has been shown to be involved in resistance to chemotherapeutic drugs in many forms of cancers. Recent efforts from the Abbott Laboratories resulted in the development of the acylsulfonamide compound and clinical candidate that targets selectively Bcl-2, Bcl-x_L, and Bcl-w while it is not active against Mcl-1 and Bfl-1. However, early clinical and preclinical studies suggest that pan-Bcl-2 antagonists, targeting simultaneously Mcl-1, Bcl-xL, and possibly all other four antiapoptotic Bcl-2 proteins, may result in more efficacious drugs. Here, following an NMR fragment-based approach, SAR by ILOEs, we report on compounds that exhibit nanomolar affinities for both Bcl-x_L and Mcl-1 in vitro. We believe that these molecules can be used as useful starting point for the development of novel Bcl-2 antagonists, in particular targeting Mcl-1.

We are motivated by the need for a generic object proposal generation algorithm which achieves good balance between object detection recall, proposal localization quality and computational efficiency. We propose a novel object proposal algorithm, BING++, which inherits the virtue of good computational efficiency of BING [1] but significantly improves its proposal localization quality. At high level we formulate the problem of object proposal generation from a novel probabilistic perspective, based on which our BING++ manages to improve the localization quality by employing edges and segments to estimate object boundaries and update the proposals sequentially. We propose learning the parameters efficiently by searching for approximate solutions in a quantized parameter space for complexity reduction. We demonstrate the generalization of BING++ with the same fixed parameters across different object classes and datasets. Empirically our BING++ can run at half speed of BING on CPU, but significantly improve the localization quality by 18.5% and 16.7% on both VOC2007 and Microhsoft COCO datasets, respectively. Compared with other state-of-the-art approaches, BING++ can achieve comparable performance, but run significantly faster.

A point cloud is a discrete set of data points sampled from a 3D geometric surface. Chamfer distance (CD) is a popular metric and training loss to measure the distances between point clouds, but also well known to be sensitive to outliers. We propose InfoCD, a novel contrastive Chamfer distance loss, and learn to spread the matched points to better align the distributions of point clouds. As such InfoCD leads to an improved surface similarity metric. We show that minimizing InfoCD is equivalent to maximizing a lower bound of the mutual information between the underlying geometric surfaces represented by the point clouds, leading to a regularized CD metric which is robust and computationally efficient for deep learning. We conduct comprehensive experiments for point cloud completion using InfoCD and observe significant improvements consistently over all the popular baseline networks trained with CD-based losses, leading to new state-of-the-art results on several benchmark datasets. Demo code is available at https://github.com/Zhang-VISLab/NeurIPS2023-InfoCD.

Cross-modal vehicle localization is an important task for automated driving systems. This research proposes a novel approach based on LiDAR point clouds and OpenStreetMaps (OSM) via a constrained particle filter, which significantly improves the vehicle localization accuracy. The OSM modality provides not only a platform to generate simulated point cloud images, but also geometrical constraints (e.g., roads) to improve the particle filter’s final result. The proposed approach is deterministic without any learning component or need for labelled data. Evaluated by using the KITTI dataset, it achieves accurate vehicle pose tracking with a position error of less than 3 m when considering the mean error across all the sequences. This method shows state-of-the-art accuracy when compared with the existing methods based on OSM or satellite maps.

In this paper, we propose a novel image similarity learning approach based on Probabilistic Feature Matching (PFM).We consider the matching process as the bipartite graph matching problem, and define the image similarity as the inner product of the feature similarities and their corresponding matching probabilities, which are learned by optimizing a quadratic formulation. Further, we prove that the image similarity and the sparsity of the learned matching probability distribution will decrease monotonically with the increase of parameter C in the quadratic formulation where C ≥ 0 is a pre-defined data-dependent constant to control the sparsity of the distribution of a feature matching probability. Essentially, our approach is the generalization of a family of similarity matching approaches. We test our approach on Graz datasets for object recognition, and achieve 89.4% on Graz-01 and 87.4% on Graz-02, respectively on average, which outperform the state-of-the-art.