AI camera traps: Wildlife Insights vs Agouti vs TrapTagger

Turn camera files into useful wildlife records

Camera traps generate many empty frames and repeated sightings before researchers obtain a usable species dataset. Wildlife Insights, Agouti, and TrapTagger combine AI suggestions with data organization and review, but differ in supported media, annotation structure, and access terms. The decision here concerns photographs and camera-trap video, rather than acoustic recordings or satellite imagery. The most useful product is the one that fits how the research team records, reviews, and exports observations.

Wildlife Insights

Wildlife Insights provides cloud-based camera-trap image management with AI blank filtering and species suggestions. It connects those image-processing steps to project organization, collaborative review, and a wider environment for exploring or sharing wildlife data. That makes it relevant to conservation programs whose goal extends beyond labeling files on one researcher’s computer. Approved data contributors can upload projects, while public-access users explore available records with restrictions around sensitive information. The product’s treatment of sharing and embargoes should be considered part of the workflow, rather than an incidental storage setting. Its current FAQ says that video is not processed directly; extracting image sequences is a separate step. Free access is also based on user and organization eligibility, with fees for some organizations. Wildlife Insights therefore suits teams seeking a connected image-data ecosystem, provided its contribution, media, and commercial terms fit the project.

Agouti

Agouti organizes camera deployments, image sequences, and observations in a research-oriented web workflow. Teams select an appropriate AI model in project settings and send deployments for automatic annotation. The resulting observations show model and confidence information, and human annotators can inspect sequences, correct species or counts, and add contextual information. Its model choices distinguish broad blank/animal/human detection from the regional species classification that follows. This structure is valuable when multiple annotators need to follow one field protocol across many cameras and sampling periods. Export uses the Camera Trap Data Package format, with downstream tools and a public API supporting further research. Data is hosted on Wageningen University and Research servers in the Netherlands. The main fit is a coordinated scientific dataset rather than a casual gallery of wildlife photographs. Because Agouti is centralized and model coverage varies, connectivity and an appropriate species model remain practical constraints.

TrapTagger

TrapTagger combines animal detection and regional species classification with human annotation for camera-trap images and videos. Its scope extends to assisted individual identification: coat-pattern similarity suggestions help researchers build a database of known animals, with people confirming the matches. That is a useful distinction for studies of recognizable patterned species, because identifying an individual is a different task from identifying its species. Additional tools address timestamp extraction, parallel review, and built-in analysis, while integrations can connect selected sightings to systems such as EarthRanger or WildBook. The platform is described as free, with organization registrations manually reviewed. Its value is strongest when media types or individual matching would otherwise require several workflows. Regional classifiers and visible coat patterns still determine the useful automation; the system does not establish reliable identity for every species. Researchers retain responsibility for confirming ambiguous labels and suggested matches.

Compare media and the unit of review

Wildlife Insights is an image workflow and requires a separate conversion step for video. Agouti centers annotation on deployments and sequences, with image-level predictions contributing to reviewed observations. TrapTagger explicitly accepts images and videos and offers several annotation routes. Wildlife Insights is attractive when a program already works with still photographs and wants connected project sharing. Agouti’s sequence structure is useful when consistent observation units matter to a research protocol. TrapTagger is the clearer candidate when video is a substantial part of the collection. These workflow differences can matter more than comparing uncalibrated model-accuracy headlines.

Compare species labels and individual identities

All three help turn large camera collections into species records, but their additional emphasis differs. Wildlife Insights links classification with a broader wildlife-data ecosystem. Agouti exposes model choices and confidence in a structured annotation process. TrapTagger adds assisted coat-pattern matching for known individuals. A regional species model is not interchangeable with an individual-recognition workflow: the former asks what animal is present, while the latter asks whether it is the same recognizable animal seen before. For a population study using individual encounters, TrapTagger’s matching tools may therefore change the shortlist. For standardized species observations, Agouti’s sequence and export structure may matter more.

Compare access and data workflows

Wildlife Insights offers free eligibility for several user groups but expects annual fees from some organizations. Agouti is free for academic and non-profit use, with commercial or large projects requiring discussion of contributions; its 250 GB guidance is not a hard universal storage ceiling. TrapTagger advertises free platform use and manually reviews registrations. These terms should not be reduced to three identical free plans. Wildlife Insights combines contribution and sharing choices, Agouti supports standardized research exports, and TrapTagger connects review to media and individual-matching tools. Choose the access model together with the team’s intended research workflow and the data it needs to produce.

Which should you choose?

Consider Wildlife Insights for shared camera-image projects, Agouti for coordinated sequence annotation and Camtrap DP exports, or TrapTagger for video and assisted individual matching. Match the platform and regional model to the study’s media and observation needs.