In this paper, we design experiments to verify such misfit between global re-ID feature distances and local matching in tracking, and propose a simple yet effective approach to adapt affinity estimations to corresponding matching scopes in MTMCT.
Because of the expensive data collection process, micro-expression (MiE) datasets are generally much smaller in scale than those in other computer vision fields, rendering large-scale training less feasible.
We show how most tracking tasks can be solved within this framework, and that the same appearance model can be successfully used to obtain results that are competitive against specialised methods for most of the tasks considered.
Ranked #1 on Video Object Segmentation on DAVIS 2017
We show that compared with real data, association knowledge obtained from synthetic data can achieve very similar performance on real-world test sets without domain adaption techniques.
This paper proposes a self-supervised learning method for the person re-identification (re-ID) problem, where existing unsupervised methods usually rely on pseudo labels, such as those from video tracklets or clustering.
This paper provides a pair similarity optimization viewpoint on deep feature learning, aiming to maximize the within-class similarity $s_p$ and minimize the between-class similarity $s_n$.
Ranked #1 on Face Verification on IJB-C (dataset metric)
Due to the continuity of target trajectories, tracking systems usually restrict their data association within a local neighborhood.
In this paper, we propose an MOT system that allows target detection and appearance embedding to be learned in a shared model.
Ranked #12 on Multi-Object Tracking on MOT16 (using extra training data)
The softmax loss and its variants are widely used as objectives for embedding learning, especially in applications like face recognition.
The key idea is that we find the local context in the feature space around an instance (face) contains rich information about the linkage relationship between this instance and its neighbors.
That is to say, for some queries, a feature may be neither discriminative nor complementary to existing ones, while for other queries, the feature suffices.
In our vehicle ReID framework, an orientation invariant feature embedding module and a spatial-temporal regularization module are proposed.