{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/directional-statistics-based-deep-metric","title":"Directional Statistics-based Deep Metric Learning for Image Classification and Retrieval","arxiv_id":"1802.09662","date":"2018-02-27","proceeding":null,"authors":["Xuefei Zhe","Shifeng Chen","Hong Yan"],"abstract":"Deep distance metric learning (DDML), which is proposed to learn image\nsimilarity metrics in an end-to-end manner based on the convolution neural\nnetwork, has achieved encouraging results in many computer vision\ntasks.$L2$-normalization in the embedding space has been used to improve the\nperformance of several DDML methods. However, the commonly used Euclidean\ndistance is no longer an accurate metric for $L2$-normalized embedding space,\ni.e., a hyper-sphere. Another challenge of current DDML methods is that their\nloss functions are usually based on rigid data formats, such as the triplet\ntuple. Thus, an extra process is needed to prepare data in specific formats. In\naddition, their losses are obtained from a limited number of samples, which\nleads to a lack of the global view of the embedding space. In this paper, we\nreplace the Euclidean distance with the cosine similarity to better utilize the\n$L2$-normalization, which is able to attenuate the curse of dimensionality.\nMore specifically, a novel loss function based on the von Mises-Fisher\ndistribution is proposed to learn a compact hyper-spherical embedding space.\nMoreover, a new efficient learning algorithm is developed to better capture the\nglobal structure of the embedding space. Experiments for both classification\nand retrieval tasks on several standard datasets show that our method achieves\nstate-of-the-art performance with a simpler training procedure. Furthermore, we\ndemonstrate that, even with a small number of convolutional layers, our model\ncan still obtain significantly better classification performance than the\nwidely used softmax loss.","url_abs":"http://arxiv.org/abs/1802.09662v2","url_pdf":"http://arxiv.org/pdf/1802.09662v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"directional-statistics-based-deep-metric","repo_url":"https://github.com/Abdelhamid-bouzid/Deep-metric-learning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"metric-learning","task_name":"Metric Learning"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":null,"task_name":"Triplet"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1802.09662","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1802.09662"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Abdelhamid-bouzid/Deep-metric-learning","reach":null}],"summary":{"ran_fixture":1},"by_repo_kind":{"listed":{"samples":1,"ran":1,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":1,"samples":[{"code_sha256_prefix":"43243f863383c540","entry":"label_mat","repo":"Abdelhamid-bouzid/Deep-metric-learning","repo_kind":"listed","path":"Triplet/compute_triplets.py","file_url":"https://github.com/Abdelhamid-bouzid/Deep-metric-learning/blob/HEAD/Triplet/compute_triplets.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"43243f863383c540"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}