Color Invariants for Person Reidentification

Abstract

We revisit the problem of specific object recognition using color distributions. In some applications--such as specific person identification--it is highly likely that the color distributions will be multimodal and hence contain a special structure. Although the color distribution changes under different lighting conditions, some aspects of its structure turn out to be invariants. We refer to this structure as an intradistribution structure, and show that it is invariant under a wide range of imaging conditions while being discriminative enough to be practical. Our signature uses shape context descriptors to represent the intradistribution structure. Assuming the widely used diagonal model, we validate that our signature is invariant under certain illumination changes. Experimentally, we use color information as the only cue to obtain good recognition performance on publicly available databases covering both indoor and outdoor conditions. Combining our approach with the complementary covariance descriptor, we demonstrate results exceeding the state-of-the-art performance on the challenging VIPeR and CAVIAR4REID databases.

Salient Color Names for Person Re-identification

“…[13]. Kviatkovsky et al [13] propose a novel illumination-invariant feature representation based on logchromaticity (log) color space and demonstrate that color as a single cue has a relatively good performance in identifying persons under greatly varying imaging conditions.…”

“…[1,4,18,15,13,31,12,7]), which is our main concern in this paper and (2) person matching (e.g. [30,19,28,6,11,5,32,14]).…”

“…But color information, in comparison with texture information, seems to be a more important cue due to the fact that in most cases, only low-resolution images can be obtained. Traditional color information such as color histogram, a simple yet effective feature representation, is most widely used in [1,4,18,15,13,31,12,19,6,11,5,32]. With the consideration of the influence of illumination variations, we calculate color histograms in different color spaces separately and fuse them to make the final feature more robust to illumination changes.…”

Multiple-Shot People Re-Identification by Patch-Wise Learning

“…It represents the expectation of obtaining correct match in top n candidates. As in most people re-identification studies [2], [4], [5], [13], we use CMC curves to show the performance of re-identification. The area under the curve (AUC) and proportion of uncertain removed (PUR) [26] are also employed to accommodate the information of performance across the CMC curves.…”

“…Such a procedure can be conducted by employing multiple-shot approaches (i.e., using multiple images) or single-shot approaches (i.e., using only one image). Single-shot approaches have been well studied during the last few years [2]- [6]. On the other hand, multiple-shot approaches, which are more practically applicable when long-term observations are available, have attracted increasing attention recently [7]- [11].…”