Research map / Detection and segmentation / Tracking
Transparent Object Tracking Benchmark
Heng Fan, Halady Akhilesha Miththanthaya, Harshit, Siranjiv Ramana Rajan, Xiaoqiong Liu, Zhilin Zou, Yuewei Lin, Haibin Ling
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add transparency feature branch to ATOM (ablation) | ATOM · TOTB | SUC: 0.614 -> 0.641 | Improved |
| use transparency feature alone instead of visual feature (ablation) | ATOM · TOTB | SUC: 0.614 -> 0.625 | Improved |
| add visual feature to transparency-only tracker (ablation) | TransATOM-V · TOTB | SUC: 0.625 -> 0.641 | Improved |
| add transparency feature to DiMP | DiMP · TOTB | SUC: 0.594 -> 0.613 | Improved |
| add transparency feature to KYS | KYS · TOTB | SUC: 0.597 -> 0.619 | Improved |
| switch backbone from ResNet-18 to ResNet-50 (ablation) | ATOM · TOTB | SUC: 0.614 -> 0.608 | Hurt |
| switch backbone from ResNet-18 to ResNet-50 (ablation) | DiMP · TOTB | SUC: 0.605 -> 0.594 | Hurt |
| switch backbone from ResNet-18 to ResNet-50 (ablation) | PrDiMP · TOTB | SUC: 0.639 -> 0.633 | Hurt |
| switch backbone from ResNet-18 to ResNet-50 (ablation) | SiamRPN++ · TOTB | SUC: 0.585 -> 0.617 | Improved |
| switch backbone from ResNet-18 to ResNet-50 (ablation) | TransATOM · TOTB | SUC: 0.641 -> 0.632 | Hurt |
| replace ATOM with TransATOM on partial occlusion | ATOM · TOTB (partial occlusion) | SUC: 0.611 -> 0.635 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- ATOM — ATOM: Accurate Tracking by Overlap Maximization (CVPR 2019) · better on 8 of 10 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- LaSOT: A High-quality Large-scale Single Object Tracking Benchmark (IJCV 2020 · similarity 0.91)
- TAO: A Large-Scale Benchmark for Tracking Any Object (ECCV 2020 · similarity 0.91)
- A New Dataset and a Distractor-Aware Architecture for Transparent Object Tracking (IJCV 2024 · similarity 0.91)
- OVT-B: A New Large-Scale Benchmark for Open-Vocabulary Multi-Object Tracking (NeurIPS 2024 · similarity 0.90)
- VastTrack: Vast Category Visual Object Tracking (NeurIPS 2024 · similarity 0.89)
- DiffusionTrack: Diffusion Model for Multi-Object Tracking (AAAI 2024 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
