Dacheng Tao, Hang Yu, Jing Zhang, Yufei Xu, Wei Zhao, Ziyu Guan
We lifted 9 functions out of this paper's own repositories and ran 9 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| rikichou/mmpose | pwc_unofficial | 7 of 7 |
| vitae-transformer/aptv2 | pwc_unofficial | 2 of 2 |
| Function | Status | Where it lives |
|---|---|---|
| aggregate_stage_flip | Ran | rikichou/mmpose/mmpose/core/evaluation/bottom_up_eval.py code served (permissive licence) · get_code("4670b554f85bd5fd") |
| compute_similarity_transform | Ran | rikichou/mmpose/mmpose/core/evaluation/mesh_eval.py code served (permissive licence) · get_code("125c410fd33efbc4") |
| covert_keypoint_definition | Ran | vitae-transformer/aptv2/demo/body3d_two_stage_video_demo.py code served (permissive licence) · get_code("10e3dd78691455b0") |
| extract_pose_sequence | Ran | rikichou/mmpose/mmpose/apis/inference_3d.py code served (permissive licence) · get_code("79272b7e8ff21207") |
| flip_feature_maps | Ran | rikichou/mmpose/mmpose/core/evaluation/bottom_up_eval.py code served (permissive licence) · get_code("a94052dc985dbd44") |
| process_face_det_results | Ran | vitae-transformer/aptv2/demo/face_img_demo.py code served (permissive licence) · get_code("000768447d5adf6c") |
| split_ae_outputs | Ran | rikichou/mmpose/mmpose/core/evaluation/bottom_up_eval.py code served (permissive licence) · get_code("43e4fca5256ad4e2") |
| vis_3d_pose_result | Ran | rikichou/mmpose/mmpose/apis/inference_3d.py code served (permissive licence) · get_code("bd6d6a2a69c4bcc1") |
| vis_pose_tracking_result | Ran | rikichou/mmpose/mmpose/apis/inference_tracking.py code served (permissive licence) · get_code("83b319f3508e6fe2") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Accurate animal pose estimation is an essential step towards understanding animal behavior, and can potentially benefit many downstream applications, such as wildlife conservation. Previous works of animal pose estimation only focus on specific animals while ignoring the diversity of animal species, limiting their generalization ability. In this paper, we propose AP-10K, the first large-scale benchmark for mammal animal pose estimation, to facilitate research in animal pose estimation. AP-10K consists of 10,015 images collected and filtered from 23 animal families and 54 species following the taxonomic rank and high-quality keypoint annotations labeled and checked manually. Based on AP-10K, we benchmark representative pose estimation models on the following three tracks: (1) supervised learning for animal pose estimation, (2) cross-domain transfer learning from human pose estimation to animal pose estimation, and (3) intra-and inter-family domain generalization for unseen animals. The experimental results provide sound empirical evidence on the superiority of learning from diverse animals species in terms of both accuracy and generalization ability. It opens new directions for facilitating future research in animal pose estimation. AP-10k is publicly available at https://github.com/AlexTheBad/AP10K 3 . * Equal contribution. The work was done during the first authors' internship at JD Explore Academy. † Corresponding author 3 The code will also be integrated into mmpose. 35th Conference on Neural Information Processing Systems (NeurIPS 2021) Track on Datasets and Benchmarks.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2108.12617")
get_code_for_paper("2108.12617")
have("2108.12617")
Connect an agent — have() is free.